从草蜢测试到星舰打捞：SpaceX 全可重复使用火箭的演进之路

Elon Musk / Narrator: It is a very old idea that the holy grail of rocketry is a 100% reusable rocket. And for whatever reason, people started to think it wasn't possible and gave up on reusability. But it's reusability that unlocks the solar system for us. If we can show that full reusability is possible, anything is possible in terms of space exploration.

Mission Control (Controller A): Wow, that's all plasma so far. Temperatures are coming up, but we got a long way to go. Anything on the inside?

Mission Control (Controller B): All clean. We've got nothing on the outskirt or attic areas. No glow in the tanks. No glow in the O drive.

Mission Control (Controller A): All right, flaps are about to take over. Flaps have taken control of the vehicle.

Commentator / Engineer: Starship is in the peak heating phase of entry. Now we get serious. When the rocket is re-entering the atmosphere at 25 times the speed of sound, every air molecule is hitting it like a bullet. That energy's got to go somewhere and where it goes is into heating up the rocket. Starship is at maximum entry dynamic pressure.

Mission Control (Controller C): We just got drone 2 live stream. We are on. All cameras are go.

Commentator / Engineer: Starship is executing the swoosh maneuver. Dynamic pressure is coming down. Starship has landed.

Team Member A: No way. It's going. We got a whole rocket. We got the whole thing. It looks like the ship is going to give us some time.

Team Member B: Can we go get those things?

Team Member C: We had a plan to do that and we have descoped it for the last two flights for V3.

Team Member B: What's the closest like ship or plane or whatever?

Team Member C: Well, we have a ship fairly close that is very much it's big enough to ridiculous.

Team Member A: Someone suggested that the flaps paddle us home.

Team Member B: Just start swimming.

Team Lead: We're going to start getting scrappy and see what we got. I mean, we don't have the toy.

Team Member B: Get it back.

Field Engineer: We will try. This is the absolute gold of the data. And for some reason, we don't understand. The rocket is just there floating and happy. We don't know how long that's going to last.

Early R&D Engineer A: Grasshopper was the experimental test vehicle that we used to learn how to do vertical takeoff and landing.

Early R&D Engineer B: The first Grasshopper flights were these very, very short hops. We called them baby hop and bunny hop was just barely off the ground.

Launch Control: 1 z launch closing.

Early R&D Engineer A: The Grasshopper team was really the SpaceX within SpaceX. We were the folks doing the stuff that was kind of out there with a very small team doing stuff that we weren't really trained to do. returning a booster, it becomes this question of like how do you get through all the stages of aerodynamics? How do you get through all the stages of flight? And the very last piece is the landing phase. And that's really what Grasshopper was about. And at the time, landing something this size after coming back from orbit seemed completely crazy. Convincing the people of SpaceX that this was possible was like the first step because once we had the whole engineering effort of the company behind us, then basically it's make a list of all the things that have to get done and go knock that list down until you have the rocket. After the Grasshopper program had shown that vertical takeoff and landing was possible, we now had to prove that we could take an orbital rocket all the way to space and land it. Proving that reusability for an operational rocket was possible. You want to go into a launch with 90% of the stuff figured out, but there's a final 10% that you really can't figure out unless you try. We had three landing attempts in a row where we had unexpected failures, but we were getting closer.

Narrator: CRS7 was a customer mission for NASA. CRS is the crew resupply.

Launch Announcer: We have liftoff of the Falcon 9.

Senior Systems Engineer A: NASA had really taken a risk on SpaceX at the time. The early days of SpaceX really NASA was a huge piece of how we were able to survive. Acknowledging a lot of warnings here.

Launch Control: We have non-nominal flight.

Range Safety: Actually, ma'am, we have missile black confirmed destruct.

Senior Systems Engineer A: It's hard to explain. That was the first failure of Falcon 9. Falcon 9 had had a perfect track record and everyone was just shocked like what happened? It was a real wakeup call for us. A lot of us that had been working reusability just went heads down because we had to fix this.

Senior Systems Engineer B: Which meant that the whole company had to shut down, figure out what went wrong, and that made the Orbcom mission not only the return to flight for the entire company, but we also decided to go for broke and bring the rocket all the way back to Cape Canaveral and land on a landing pad.

Countdown: 3 2 1 zero.

Telemetry Engineer: So, I'm staying inside. I'm watching data. It was a very tense count.

Telemetry Audio: Griffin pump is on.

Flight Controller: Stage one's an entry startup.

Propulsion Engineer: There was a couple of weird things that happened in the second stage. Well, maybe it's a helium leak. Maybe it's a nitrogen leak. Maybe it's just water. We don't know.

Controller A: Let's go on the com and ask if anyone's seen this before.

Controller B: Shannon just asked. What did they say?

Controller A: They're talking about it. We're standing there debating like, "Do we go? Do we not go? Stage one boost back is starting RF and power still look healthy.

Operations Lead: Finally, we're like we're just going to go and

Commentator / Engineer: stage one entry startup. We had a tracking camera on the back of the rocket and this video when the engine lit was one of the most amazing images I have ever seen. And like I was just completely speechless. Huge fire. Landing legs deploy. And the next thing you see, the rocket's like landing.

Mission Control Announcer: Stage one has landed.

Ground Ops: LZ1, the Falcon is landed. Landing operators movement procedure 1100 station

Team Member C: that was awesome.

Team Member D: Oh yeah.

Team Member C: Awesome.

Team Member E: Absolutely not. is not a

Senior Engineer: maybe like we're not sure and then all the great all the octopus bay doors are gripping around but I think it's just conversation with them pros to this day was like an amazing moment that was basically the peak of everything like everything we've been working on came to that day because we still like even though we've been landing rockets and trying to land them on the drone ship we hadn't successfully landed anything. There's no sit at all where the landing legs were flush against the side of the rocket. So,

Team Member: yeah.

Colleague: Why are you guys still at work?

Flight Performance Lead: And that is the rocket right there. What we did over the next year or two is successively get more and more performance out of the rocket. Basically, the shorter you can make the landing burn, the less payload you take away from the capacity of the rocket. So to do that, we had to keep pushing the envelope. With Falcon, we achieved partial reusability, but we showed that full reusability might be possible, and Starship is what's going to achieve that. We're 500 to 600 nautical miles off the coast of Western Australia. Pretty much as remote as you can get.

Marine Recovery Lead: So, this is the first time that we've ever hooked up to the Starship rocket. It's right there off our stern side.

Salvage Crew Lead: And together, we're going to bring this puppy home. recommend astral to move 20 mort.

Crew Member: We're all set.

Ocean Engineer: Starship is it's actually a bad boat, but it's an amazing buoy. It has an enormous amount of buoyancy for how large it is. So, its draft is very, very small. You're talking like a foot and a half, 2 feet of draft in the water, and it has a lot of sail area. So, with just the slightest amount of proddding or the slightest amount of wind, it just kind of wanted to go do whatever it wanted to do. No, we've lost it.

Tug Captain: Just lay down a bit, mate. Just slow down a bit. Right. Just keep going out nice and easy. You're starting to get too far.

Field Operator: Yeah, we're struggling a bit. Okay, that's good.

Maritime Crew: It fought us the entire way. The rocket didn't want to get recovered. It seemed like it just wanted to float out in the ocean forever.

Expedition Lead: We've come to an old stop now. drag it out. Austral is doing his bit. We're doing our bit. It's been a pretty crazy couple weeks. Things have not gone to plan. I think it's safe to say the weather started kind of turning against us. We got to be like a 16 to 18 foot seas with about 40 to 45 knot winds and it just started this vicious cycle of fighting tooth thin nail for like the teensiest tiniest smallest victories. We'd make the littlest hint of progress and then bam, again, like in the course of an hour, it would all kind of fall apart at the end of the day. We were still like 850 plus miles from Dampier. And basically said, hey, this thing's never making it back to mainland Australia. At this point, it's sort of like we have to pivot or we're not getting this back at all. The real problem was that we were making extremely slow tow essentially like just barely fighting the drift. And even at that speed, we were starting to damage the vehicle pretty hard. The thing that ended up working was just this sho string tied to the back of D1 and we were like jerking this thing along essentially for 400 miles of the ocean with another like 200 mile assist of just the waves happen to be in the right direction. We certainly learned a lot which is hopefully knowledge we'll never need again.

Team Member A: Wow. You guys ever think you'd see something like this?

Team Member B: No,

Team Member C: I never did.

Team Member A: We found a starship.

Field Engineer A: Some good looking tiles.

TPS Lead: You guys must have glued them on there good and tight. So the fundamental challenge with returning the Starship second stage is how do you make it survive the extreme heating of entry. So we put a lot of work into the heat shield to handle as much heat as possible. But also the first job of the heat shield tiles is to survive a set. Right at liftoff, the vibration from the engines and the acoustics of the engine plume bouncing back from the ground is extremely violent. Then as the rocket ascends, it gets faster and faster. And before the air gets thin, there's a period of maximum dynamic pressure. The Starship second stage looks fundamentally different from any second stage you've seen. And that's because unlike Falcon, we started from a clean sheet design. If we tried to land the rocket the same way that we land the Falcon booster, we would just melt up the engines on the way in, no amount of material science could save us. We wouldn't have a rocket by the time it got to the ground because the best way to test tiles in particular is flight. Often times it is a try something, see if it works, and then iterate from there. Starship is an exercise in iteration and the tiles in particular are just a symphony of iteration. Heat. Heat. Heat. Heat. Yeah, these are cool. It's kind of funny. This tile represents almost half a decade, if not more, of development effort.

TPS Engineer A: They always tell you not to reinvent the wheel, but basically we had to reinvent the wheel on the entire TPS program for Starship because this is the largest orbital vehicle ever built. You have to think of plasma or the heat load during re-entry like a river. If the heat was coming in perpendicular, none of this would matter cuz all these tiles would melt away. But since this heat load is flowing down, it doesn't have enough time to destroy the coating of the tile. We're trying to prevent the heat from getting from the surface of the tile to the structure of the rocket.

TPS Engineer B: And so the trick you play is that the white material is an incredibly good insulator. You could take the tile and just blast a torch in the middle of it and hold it like this for 10 minutes and not feel a thing. The heat shield tile is essentially just really fancy glass. One of the key challenges in general with the heat shield is there are about 20,000 tiles on the ship, which means that you have 20,000 opportunities to get it wrong. And so the act of actually attaching and installing the tile is one that has to be undertaken quite carefully to make sure that every single time every tile is fully installed and attached.

Starbase Technician: Our job is it's detailed. You know, we we got to pay attention to the little twins. We look at millimeters. I'm a local. I'm from the valley. I started here. You know, it makes me proud, you know, like saying that we're from here, home of a starship. We came to work one day, all of a sudden they asked, "Hey, you got your passport?" I'm like, "Yeah, I do." Like, "All right, prepare. You're going to Australia."

Recovery Engineer: It felt like one day to another. One day I was here on Starbase. The other day I woke up next to Starship in the Indian Ocean. Pretty nice. Look.

Technician A: Yeah. Look at that.

Technician B: Dang.

Technician A: There's your glowing right there.

Salvage Director: One step closer to make it fully reusable. This is top five mornings at work ever. People in scuba, people pulling tiles. We're at Christmas Island. It's very, very good. The current configuration of the vehicle does not allow us to do inspections in all the places we want. So, one of our main goals today is to roll the vehicle.

Rigging Engineer: Yeah. That rigging will be underneath the rocket. And as we start to inflate the lift bag on this side, it's going to slightly roll the vehicle up.

Ops Coordinator: Give me a hand signal when it looks like it's starting to take shape.

Observer: Yeah, she's rolling.

Rigging Engineer: Wait, wait, wait, wait, wait.

Technician: TBS guys are going to love this.

Salvage Director: Yeah.

Expedition Lead: Ultimately, what we're doing has never been done before. And so, it's really critical that we have a multitude of skills in order to meet the demands of the job. Ladder coming up. All right. All right. We've had almost 30 SpaceX employees come to the Indian Ocean. Those 30 employees have been from across the entire company. Whether that is EHS, whether that's recovery, we have a ship integration engineer out here. We have a ship structures engineer. We have a naval architect. We have a fluids engineer tile team. Of course,

Chief Structural & Recovery Engineer: my first reaction when I saw the rocket tip over and survive was like of amazement because I was not expecting that and I was like rejoicing. And the second thought was like, oh, somewhere in the Indian Ocean, we are sitting on a mine of gold, of data that would allow us to improve the design at least 3 months earlier than what I thought we were going to be able to do. I'm going to bring you some more pieces. We have a chance to get the data to fix the new one and figure out what's the next layer of the onion that we need to unwrap and to make the vehicle fully reusable. Tomorrow we're going to start very early so that we have plenty of daylight and we will basically float ship 40 on to the deck of the Forte which is a heavy lift ship. We will secure it above a cradle which will support it and we will debalance the Forte and ship 40 will set in the cradle where we can secure it down and make sure that it's safe and secure for a trip eventually back to Starbase. When ship 40 survived, we quickly mobilized Forte, came up with a plan, designed this cradle, and we actually assembled the cradle as we were sailing from uh from Singapore over here to Christmas Island. She's scrappy, but she's really robust. 2 m. Okay. Shall we come closer to the guide post and then slide it upwards? This is good. If you can just keep the tension like this on the line. Very good.

Crane Operator: Now we just start raising this thing.

Engineering & Recovery Director: At this time uh we are getting the vehicle S40 directly above the cradle. And then once teams confirm that the vehicle is stable and steady, we will decide to debalance the Forte. allow all of the bed. It's been just over a month since this rocket flew and uh it's been 2 days since we got it on this barge dry. I think in terms of the data we've been able to get back, this mission has been an absolute success. Every single one of the discoveries we made on ship 40 will get rolled into improvements for the heat shield on ship 41. And the reason that this is so critical is that the next big goal for the program is to return the ship to the launch site. And to do that, we need to rely on our heat shield more than we ever have in the past. I think the core potential that we have is to redefine the way that humans interact with space. Space is something for all of us, for humans in general. And it is something that I think we are on the precipice of unlocking with a redefinition of how difficult it is to go from here on the ground to low Earth orbit. The really big news from this flight is that this was the first time on the end of the mission in the middle of the ocean. We set it up as if it was doing a tower catch. So, we had all of the landing guidance algorithms go through the maneuvers to get the extreme precision that you would need to put the rocket over the top of the tower and into the arms. And it did that. It achieved that. So this is the first time that we can say that had there been a tower out there, the rocket would have been caught.