SpaceX's Crazy Plan On Starship Double Catch Shocked NASA Engineers

We will try to actually catch the first and the second stage of Starlink V3 this year, and we could possibly catch the ship as soon as the next flight. Wow! Did you hear that? After years of speculation, Elon Musk has finally revealed when SpaceX plans to catch both stages of Starship. So, how is the company preparing for such a risky mission? And while the ship's next step is coming into focus, one major question remains unanswered. When will SpaceX attempt the first catch of the V3 Super Heavy booster? For years, the single biggest question mark hanging over Starship wasn't whether it could reach orbit. It was whether the ship could survive coming back. Reentry is brutal. A spacecraft screaming through the atmosphere at hypersonic speed generates enough heat to melt most metals, and Starship's upper stage, simply called ship, has to shrug that off using a layer of ceramic tiles bolted to its skin. Get that wrong, and you don't get a rocket back. You get debris. On July 24th, ship made it through. Flight 13 sent the vehicle on a suborbital hop that ended with a soft and very precise splashdown in the Indian Ocean, off the coast of Western Australia. The heat shield held. So did the rest of the vehicle, which stayed floating and intact for days afterward, giving engineers a rare chance to recover it and study exactly how it fared. During SpaceX's earnings call this week, CEO Elon Musk didn't hide his relief. But all indications from data and our visual inspection suggest that we have solved the heat shield problem, which is arguably the single biggest problem. That doesn't mean we won't make improvements to the heat shield. That's a notable shift in tone. For most of Starship's development, the heat shield had been the program's most stubborn puzzle, the kind of problem that swallows years of testing and still leaves room for surprises. However, heat shield survival, controlled descent, and soft splashdown are only a small part of the story. Flight 13 goes further with the successful separation of stages and reactivation of one of the Raptor engines while in space. Put together, those are core capabilities necessary to achieve return to Starbase for catch. That clears the way for the next and far more dramatic step. It's to catch the ship, called S-41, immediately in Flight 14. It's the one thing the company hasn't tried yet, and it's exactly what space enthusiasts have been circling on their calendars. Here's why it's such a leap. SpaceX has already caught the super-heavy booster three times using the tower's mechanical arms, nicknamed Chopsticks. But the booster is the easier half of the job. It comes back down from a relatively low altitude and a slower speed. The ship is a different animal entirely. It returns from orbital velocity, meaning it's moving several times faster, and it has to thread the needle on its final approach with almost no room for error. Catching the booster proved the arms work. Catching the ship proves the whole system works. To pull this off, crews have spent the weeks since Flight 13 quietly getting the launch tower ready. Inspections and repairs began almost immediately after that flight. And the good news is the pad itself held up well. The concrete surface and the surrounding berm walls came through mostly intact, which isn't something you can take for granted when 33 engines light up a few feet away. Beyond that, there's been quieter work happening on the tower's plumbing. The quick disconnect systems, the fittings that fuel and drain the rocket while it's stacked on the pad, appear to be getting reworked. Bundles of separate pipes seem to be getting swapped out for a single consolidated line, based on hardware spotted being lifted on and off the tower's top deck. SpaceX hasn't confirmed exactly what this is for, but the timing and location point toward the fueling connections tied to the catch sequence. Engineers have also been running mechanical tests on the booster's hold-down clamps, extending and retracting them to make sure everything that has to grip, release, and grip again performs exactly as expected when it matters. As for how Ship 41 will actually fly its return, the smart bet is that it is expected to follow the same late-commit, dynamic banking re-entry profile that S-40 flew. A dynamic banking descent is, essentially, a controlled S-curving glide back through the atmosphere that lets the vehicle keep adjusting its approach right up until the last possible moment. The whole point of flying it this way is to preserve an escape option. If anything looks off as the ship nears the tower, it can still bail out and land safely in the ocean instead of committing to a catch it isn't ready for. That safety margin matters more on a first attempt, not less. And this time, the vehicle is coming in with more energy and from a longer trajectory. Since it's a step closer to the true orbital return, Starship will eventually need for routine tower catches. Ship 41's flight is being treated explicitly as a rehearsal for that operational profile, which is exactly why it inherits Ship 40's approach rather than trying something new. Now, keep in mind, everything I've covered here is just the latest as of when I'm putting this together. Flight 14 is still a few weeks out, and with SpaceX now a public company, they've got every incentive to be more open about their plans and progress than they used to be. So expect more details to surface as we get closer to launch. That's exactly where this channel comes in. I'm not just trying to get you the news first, though we usually do. What I actually care about is digging into the story behind the headlines, the details most channels either don't catch or don't bother explaining. That's the stuff you won't find anywhere else. So, if you don't want to miss what's coming next, hit subscribe. I've got a lot more of this breakdown lined up before Flight 14 gets here. Now let's talk about an interesting wrinkle. Back in V2, catching the booster was always SpaceX's priority. With V3, that order has effectively flipped. It's entirely possible SpaceX catches a V3 ship before it ever catches a V3 booster, even though Elon has said both are targeted for this year. The logic behind putting the ship first comes down to the heat shield. Surviving reentry has always been the program's toughest remaining problem, and getting the ship back intact means engineers can finally get their hands on the tiles and see exactly how they held up, rather than reconstructing the story from sensor data and video. That's worth more to the program right now than another booster catch would be. There's also a simpler operational reason to keep Flight 14 to a single recovery. Catching both stages on the same mission means the booster has to be caught, then cleared off the tower fast enough to make room for the ship coming in behind it, a maneuver known as fast destacking. That's a lot of choreography to get right on a flight that's already attempting a first. By catching only the ship this time, SpaceX keeps the mission focused on one new milestone instead of two. But there's a more practical reason the booster catch is on hold, and it's not really about sequencing. Both V3 stages are described internally as essentially all new vehicles compared to V1 and V2, and on Flight 13, only one of them proved it out. The ship came home clean. The booster, Booster 20, didn't. During its landing burn, the booster suffered a partial engine relight failure. Several of its engines didn't reignite as planned, which left it without enough thrust to slow down properly. It hit the gulf far harder than intended, and was destroyed on impact. That's not a minor issue to shrug off before attempting a catch-over land. So before SpaceX brings a V3 booster back to the tower, expect the same conservative path the program has followed before. Prove the landing first, over water, where a bad outcome means a rough splash instead of damage to a launch pad worth hundreds of millions of dollars. The expectation now is that Booster 21, the one slated to fly on Flight 14, needs to nail a clean, controlled splashdown in the gulf before SpaceX will even consider risking it on a tower catch. That's the same bar SpaceX held Ship 42 before clearing the ship for its Flight 14 attempt, and there's no reason to think they'll lower it for the booster. It's easy to watch rocket catches as spectacle and miss why they matter economically. Every starship that gets thrown away after one use is an enormous amount of engineering and metal that never flies again. According to estimation, the direct hardware waste is on the order of approximately 90 to 120 million dollars, with fuel and launch operations adding only a few million more. Catch it instead, and that same hardware can fly dozens or even hundreds of times, the same way an airliner does. That single change is what allows the cost of getting mass into orbit to fall dramatically, which in turn is what lets SpaceX talk about scale that sounds almost unbelievable. During the earnings call, Musk laid out just how large that scale could get. SpaceX currently delivers around 2,500 tons of cargo to orbit each year using its Falcon rockets, which by itself is already far more than the rest of the world's launch providers combined managed together. With Starship fully operational, the company's stated goal is to eventually deliver well over a million tons per year, and Musk suggested the long-term ceiling could be closer to 10 million tons annually. Numbers like that only make sense if the rocket stops being consumed on every flight. Reusability isn't a nice bonus feature for Starship. It's the entire premise the vehicle is built around. That's also why SpaceX keeps talking about flight cadence, the pace at which it launches. Musk said the company expects to be launching Starship roughly once a day within about a year, a rate that would be flatly impossible if each rocket had to be built from scratch for every mission. Catching, inspecting, refueling, and relaunching the same vehicle is the only path to that kind of tempo. Every successful catch this year is effectively a rehearsal for that future rhythm. SpaceX has set an ambitious goal for Starship. According to Elon Musk, the company wants to reach a flight rate of at least one launch every day, and possibly more, by around August 2027. That's 365 launches a year, and at that scale, building the rocket is only part of the challenge. SpaceX also has to rethink how Starship is manufactured, moved, and prepared for flight. That's where Star Factory comes in. In this video, we're going inside SpaceX's production system to see how the company is trying to turn Starship manufacturing into a process capable of supporting that kind of launch rate. I'll cover the infrastructure and logistics behind this effort in the next videos, so if there's a specific part you'd like me to cover next, let me know in the comments. And of course, don't forget to subscribe right now so you don't miss anything. But to understand why Star Factory looks the way it does today, we need to go back about a decade. In one of his early talks about manufacturing, Elon Musk described a very different way of thinking about factories. Instead of simply building a facility to manufacture a product, he argued that engineers should design the factory itself as a product. Essentially, the machine that makes the machine, or what he famously called the alien dreadnought. The true problem, the true difficulty, and where the greatest potential is, is building the machine that makes the machine. In other words, building the factory. And really thinking of the factory like a product. Not sort of a hodgepodge of things. The idea is pretty simple. To manufacture reusable rockets at high volume, it is first necessary to build machinery capable of producing rocket components with repeatability and precision, while minimizing reliance on manual labor. Jeff Bezos has expressed a similar philosophy at Blue Origin, emphasizing rate manufacturing and the urgency of building the machine that makes the machine. Rate manufacturing and driving urgency around the machine that makes the machine. However, while in Blue Origin, things are just in the beginning. SpaceX has already spent years turning that idea into a real production system. So, how does Star Factory actually work? A lot of it comes back to a manufacturing philosophy Musk developed through his experience at Tesla. It can be summed up in five steps. Step 1. Make your requirements less dumb. The first step is to question every requirement. It doesn't matter who created it. If a requirement came from a smart person, Musk argues that it can actually be more dangerous, because people are less likely to challenge it. That mindset has a major impact on how SpaceX approaches Starship production. Traditional factories often rely on a long production line. The product moves down a conveyor, and different parts are added at different stations until the finished product comes out the other end. That system has been around for more than a century, and for good reason. But Musk questioned whether a giant rocket really needs to be built that way. Instead, SpaceX uses a production approach closer to Tesla's unboxed manufacturing concept. Rather than moving one vehicle or rocket through a single linear line, major sections can be built separately and in parallel, then brought together later. For Starship, that means dividing the vehicle into major sections. The upper section includes the nose cone and payload area. The middle section contains the propellant tanks, and the bottom section contains the engine section. Each area can be worked on independently at specialized stations, while other teams are building other parts of the vehicle. And instead of sending these enormous structures down a conventional conveyor belt, Starfactory uses a flexible logistics system built around overhead cranes. The cranes move along steel beams, lifting massive rocket sections, and positioning them wherever they're needed next. Large components can also sit on rotating platforms, allowing workers and robots to stay in position while the hardware is moved around them. That may sound like a small change, but it fundamentally changes the factory. Instead of forcing the rocket to follow one fixed path, SpaceX can move the work to the people and machines doing the work. And because different sections are being manufactured at the same time, the entire process can move much faster than a purely sequential production line. There's another advantage. SpaceX can keep much of the manufacturing process under one roof. Rather than constantly moving enormous rocket components between different facilities or suppliers, major production steps can happen within the Starfactory system itself. The ultimate goal is a dramatic increase in production rate, moving from building a rocket roughly once a year toward eventually producing one every day. Once the individual components are ready, they converge at the massive Gigabyte facility for final assembly. The building stands about 116 meters tall and is designed with enough production capacity to support the simultaneous assembly of 24 Starship spacecraft and super-heavy boosters. Step 2. Delete a part or process. The second step sounds simple, but Musk considers it the hardest one. Delete anything that doesn't need to be there. Instead of asking how to make a part or process better, the first question is whether it should exist at all. Musk's rule is surprisingly aggressive. Keep deleting until you're forced to add back at least 10% of what you removed. If you never have to put anything back, you probably didn't cut enough in the first place. That philosophy has shaped the evolution of Starship. Take the super-heavy booster. With Starship version 3, SpaceX is moving from four grid fins to just three. The Raptor 3 engine also looks noticeably simpler than its predecessors. SpaceX has been removing a huge number of bolted and welded connections, including the massive flanges used around the pumps and manifolds. Every connection that disappears means fewer parts to manufacture, fewer opportunities for something to go wrong, and less work during assembly. And once SpaceX has removed everything it can, it moves to the next question. How do you make what's left simpler? Step 3. Simplify or optimize. Musk's view is that people shouldn't spend their time doing repetitive or hazardous work, especially tasks involving heavy lifting, awkward positions, or confined spaces. So instead of simply adding more workers as production increases, SpaceX has been turning many of Starship's manufacturing processes into automated operations. The change is especially obvious in the construction of the vehicle's primary structure. Machines take stainless steel from massive rolls, cut it into sheets, and form those sheets into rings roughly nine meters in diameter. Once the rings are stacked into larger sections, a robotic arm can move around the structure and complete the welds in one continuous motion. That's a major departure from the early Starship prototypes, which relied much more heavily on manual welding. SpaceX also uses hydroforming for complex curved structures such as nose cones and tank components. The process uses extremely high fluid pressure, more than 36,000 pounds per square inch, to force steel sheets into the desired shape. And automation isn't just about making the machines do the work. It's also about making sure the work is done consistently. During assembly, automated jigs hold massive components in place, while integrated sensors monitor their position. If something is slightly out of alignment, the system can detect it and make corrections before the error moves further down the production process. The same philosophy extends to Raptor engines. Raptor engines are produced on dedicated manufacturing lines where robots handle tasks such as precision welding, component transfers, and automated inspections. Only after you've questioned the requirements, deleted unnecessary parts, and optimized the remaining process should you try to make it faster. Otherwise, you're just getting better at doing something you shouldn't be doing in the first place. Musk has a blunt way of putting it. If you're digging your grave, don't dig it faster. His experience at Tesla provides a good example. During the Model 3 production ramp, Tesla tried to automate and speed up the installation of fiberglass mats designed to reduce noise and vibration around the battery pack. But the production line kept getting backed up. The obvious solution seemed to be making the machines faster. Instead, Musk eventually questioned the entire process. Did the mats even need to be there? Tesla tested Model 3 vehicles with and without the mats and used microphones to compare the results. The company reportedly found no meaningful difference in acoustic performance. So instead of making the installation process faster, Tesla eliminated it. The mats went away, along with the complicated bonding process required to install them. And that illustrates the real point of Step 4. Sometimes the fastest production line is the one that doesn't have to perform the task at all. Step 5. Automation Strategy. So why is automation the fifth step instead of the first? Because Musk believes, don't get too far along in the automation before you know what you're going to build. One of the biggest mistakes even talented engineers can make is automating a flawed process. Instead of fixing the problem, you just make the bad process run faster and much harder to change. The difficulty in making changes also links directly to the cost matter as you build that inefficiency into expensive machinery. This is where Musk's approach has some similarities to the Toyota production system. Try doing it by hand first. Implement Kaizen, continuous improvement, and eliminate waste before considering the use of machinery. In the early days, SpaceX relied heavily on tents, temporary setups, and manual welding. That may sound primitive compared with the highly automated factory SpaceX is building today. But there was a reason for it. Temporary facilities were easy to change. SpaceX could move equipment around, rearrange the workflow, and figure out which steps were actually necessary before committing them to permanent infrastructure. Once the process became more mature, SpaceX could start replacing those temporary solutions with dedicated machines and robotic systems. In other words, the robots came after the process, not before it. And that raises a much bigger question. Today's industrial robots are extremely capable, but most of them still work within fairly predictable boundaries. They follow programmed routines and perform the same operation over and over with remarkable precision. The next step is to make those machines more adaptive. SpaceX is looking toward machine learning systems that could allow robots to respond to changing conditions, learn from previous operations, and detect problems before they become defects. Imagine a welding robot that doesn't simply follow the same programmed sequence every time, but can recognize that the temperature has changed, adjust its parameters in real time, and correct the process before a human inspector ever sees a problem. That's a very different kind of factory. And this is where Tesla's Optimus humanoid robot becomes particularly interesting. Optimus is being developed around the same broader neural network and autonomous driving technology ecosystem used by Tesla's vehicles. Its intended capabilities include increasingly precise movement and manipulation, making it potentially useful for tasks that are difficult to automate with traditional fixed industrial robots. The long-term vision goes beyond the factory floor. SpaceX and Tesla have discussed the possibility of humanoid robots eventually operating in environments where sending people would be difficult or dangerous, including future work on the Moon or Mars. A robot could potentially help prepare a site, move equipment, set up solar power, or assist with habitat construction before humans even arrive.