SpaceX’s secret weapon: can withstand explosions | TECH TUESDAY special edition

Original link: https://www.latepost.com/news/dj_detail?id=1613

At 9:33 am local time on April 20, SpaceX’s Starship spacecraft slowly lifted off from a launch site in southern Texas. Nearly 4 minutes later, the spacecraft and rocket exploded together. But after the explosion, there were cheers from the SpaceX headquarters building, and someone opened a bottle of champagne.

It is already a success to make the 120-meter-high, largest rocket built by human beings escape gravity and not explode on the launch pad. And about a minute after takeoff, Starship successfully passed the point of maximum aerodynamic pressure (Max Q) – one of the most important stages in a rocket launch, meaning it withstood the pressure of breaking through the atmosphere.

2f204959636e1f6696eb0c75c8412608.gif

Starship lifts off.

It often takes a year or more for a new rocket to be ready for a second launch after a failed launch, especially after an explosive disintegration. But SpaceX’s plan is, “in a few months,” to do it again.

SpaceX has invested $3 billion into Starship, part of which is to build a production plant for Starship spacecraft and Super Heavy rockets. Earlier this year, SpaceX founder Musk said that five sets of spacecraft and rockets would be built within the year. Near the Starship launch pad stands another Super Heavy rocket.

If you count the prototype, this is Starship’s ninth explosion. This is the second time SpaceX has failed to launch a brand new rocket. Earlier, in the early days of SpaceX, the first rocket exploded three times before reaching orbit. The business company can afford to fail at a high rate and grow stronger with each failure.

21cb556d9a19d446daa94e4238d6c8cb.png

In one explosion after another, SpaceX has achieved an unprecedented technological breakthrough—the rocket landed and recovered after launch, completely changing the space industry structure dominated by government agencies. At present, SpaceX is the organization that launches the most rockets in the world every year. It helps NASA maintain the space station at a lower cost and builds a new communication system on Earth with thousands of Starlink satellites. The cost of sending a ton of equipment into space in the United States today is nearly 90% cheaper than in the late 1990s when NASA dominated.

Whether it is building infrastructure with satellites around the earth or flying to a farther future, rockets need to send heavier materials to space. The previous rocket with the largest load was the Saturn 5 built by the United States for the moon landing in the 1960s.

If Starship successfully enters orbit, it means that a commercial company has made the largest spaceship and rocket launch system in human history with 2% of the cost of the moon landing plan and transported the most materials into space at one time, and the launch cost may be low To millions of dollars, start a new stage of human space exploration.

d222ef9e70cab3cb029a3ca846c0c879.png

SpaceX’s Superpower — Poor

Compared with government agencies like NASA, SpaceX’s superpowers are poor.

NASA has realized the vision of landing on the moon promised by US President Kennedy at an astonishing speed-from Kennedy’s announcement to Congress in May 1961 to the first moon landing in 1969, it took 8 years.

The book “How Apollo Flew to the Moon” commented that the Apollo program is a model of using 20th-century technology to realize 21st-century exploration, which is two generations ahead of schedule.

During the Apollo program, the U.S. government invested in NASA without a limit—thousands of scientists and more than 400,000 workers joined it, consuming $25 billion (the currency value is equivalent to more than $150 billion today). Without powerful computers and sophisticated instruments, NASA used tedious steps to piece together the necessary conditions for landing on the moon.

Even after the end of the Cold War, NASA’s budget was greatly reduced, still $20 billion a year. SpaceX has only raised nearly 10 billion U.S. dollars in its 21 years of establishment.

96276b5674894193112a2b1d72d6a82c.png

Starship and Super Heavy are the largest rockets ever built by man. The bottom three lines are Country/Region, Altitude, and Capacity. Source: Wiki Commons

In 2022, three scholars from Oxford University analyzed 203 space missions of NASA and SpaceX, and found that SpaceX’s rocket development cost is only one-tenth of NASA’s, and the development speed is twice as fast.

They believe that the main point of departure is different. As a government agency, as long as NASA demonstrates its national technological strength by realizing advanced projects, all it does are one-off projects without scale effects, which cannot be reused and are difficult to sustain. As a commercial organization, SpaceX wants to enter space frequently with limited funds, so it must reduce costs as much as possible and complete the platform.

“Throwing away a rocket worth millions of dollars after each flight is like throwing away a Boeing 747 after each flight.” Musk believes that reusability is the key to the commercialization of the aerospace industry. The reason why there is less demand for spaceflight is that the price is ridiculously expensive. If the price is cheap, there will be demand. If there is demand, the business will be established. If the business can be established, the research and development of the entire technology will be sustainable.

This idea is not uncommon. Modern industries are all platform-based, which reduces unit construction and use costs and stimulates more demand. However, space exploration has long been planned and ordered, rather than market-led, and institutions that come from financial allocations often lack the ultimate pursuit of efficiency.

After using Falcon 1 to prove that SpaceX can launch rockets into space, Musk made rocket recovery a must-achieve goal. No institution has ever achieved such a feat before, and few can do it now.

SpaceX’s 5-year attempt to make it a reality has paid off enormously. A SpaceX spokesman said it costs SpaceX to refurbish and reuse a rocket for far less than half the cost of building a rocket from scratch. Last year, SpaceX’s Falcon 9 rocket cost $67 million per launch. The rocket project led by NASA will cost hundreds of millions or even hundreds of millions of dollars per launch. Cheap prices help SpaceX attract a lot of customers. Last year, European and American countries launched 84 rockets, more than 70% of which depended on SpaceX.

c567da73e8317ad591b920eafa5064d9.gif

SpaceX is launching 60 Starlink satellites “lined up” into orbit at a time, providing satellite internet around the world. This is a typical commercial demand that will only be available after the cost of space launches is reduced.

Musk believes that the only way to build a usable rocket is to experiment and iterate quickly to get closer to the most correct design. The Falcon 9 made 20 rocket launches before it was successfully recovered. Before the Starship test launch this time, SpaceX tested at least 10 different Starships and blew up 8 of them.

In previous space exploration, such damage was basically impossible and unaffordable. Traditional space agencies prefer to spend years and a lot of money perfecting designs to avoid explosions as much as possible before building rockets.

After repeated failures in the experimental phase, SpaceX got reliable and cheap rockets, attracting a large number of commercial orders. Last year, SpaceX launched 61 times with a success rate of 100%, so in the life cycle of a rocket, the failure rate is very low.

Rocket recovery is only one part of the cost reduction process. Just like what he did at Tesla , over the past two decades, Musk has brought the concept of extreme cost reduction to every link in the development of rockets.

With a cheaper solution, even if it is more bulky

Now the main bodies of many rockets are made of carbon fiber composite materials, which can withstand the high pressure generated when the rocket breaks through the atmosphere and are not easily deformed. More importantly, it is light enough to effectively reduce the weight of the rocket itself, allowing the rocket to carry more things or use less fuel when it rushes into space.

SpaceX tried carbon fiber composites for the shell of its Starship rocket, but gave up after building a few large parts. It is not resistant to high heat. If the temperature exceeds 200 degrees, the strength of carbon fiber composite materials will plummet. If you want to use it, you need to add a heat insulation layer, which will increase the difficulty and cost of rocket manufacturing. Moreover, carbon fiber composite materials are very expensive, with a market price of $130 per kilogram.

This runs counter to SpaceX’s philosophy of building rockets quickly and cheaply. SpaceX chose cheaper stainless steel as the main shell structure of the rocket, which Musk said was “the best design decision SpaceX made.”

The biggest problem with stainless steel is its weight. In 2019, SpaceX’s Starship prototype made of stainless steel weighs 200 tons. If it is made of carbon fiber composite materials, it will be roughly reduced to 40 tons. But stainless steel is cheap enough, about $4 per kilogram. SpaceX can save more than $4 million on a Starship by sacrificing load.

6465a505ce687045a7bee91d38a97f08.gif

Early Starship test footage. Source: SpaceX

Using heavier stainless steel, SpaceX did everything possible to reduce the weight of the rocket. Stainless steel is resistant to high temperatures. SpaceX only adds a heat shield to the part of the rocket that is in contact with the atmosphere, and does not use it elsewhere. SpaceX can also reduce some weight by not painting the rocket. Before this launch, if you zoom in to see the rocket on the launch pad, you can clearly see the welding traces. After three years of iteration, SpaceX reduced the weight of Starship from 200 tons to 120 tons.

Similar choices are made not only for the outer materials of the rocket, but also for the internal components.

Once a rocket is in space, solar radiation can cause bit flips in computer systems, such as changing from 0 to 1, causing computer processors to make mistakes. The traditional approach is to purchase computer equipment specially developed for high-radiation environments, which is extremely expensive.

Hans Koenigsmann (Hans Koenigsmann), who once served as the head of flight reliability at SpaceX, mentioned in an interview that he had the same philosophy as Musk: If it can be solved with a $500 computer, why spend $5 million? Dollar?

Each of SpaceX’s mass-produced Falcon 9 rockets is equipped with three dual-core x86 computers, and each of the two cores in each computer runs a set of flight control software. Throughout the flight, as long as one of the three computers can work normally, there will be no problems. Similarly, the Dragon spacecraft sent into space by the Falcon 9 also made a similar redundant design.

While making rockets, we are also making factories

Only hundreds of rockets are produced in the world every year, and there are even fewer rockets like Starship. The last rocket with a similar size, load, and thrust was the SLS launched by NASA last year. Only one rocket was produced in the past ten years.

But when promoting the Starship project, Musk hopes to build an efficient assembly line like the automobile industry-to ensure that parts are sent in at one end, and rockets can come out at the other end.

In order to save costs, SpaceX did not use concrete to build factories, and set up many large tents at Starbase, the base in Texas, USA. Eric Berger, a well-known American aerospace journalist, said with emotion that this place is like the US Navy Yard a few weeks after Pearl Harbor.

5401e984ef3d91f35f6dec7f1e790d70.jpeg

Starship assembly site. Source: SpaceX

Every time a rocket is produced, SpaceX iterates the pipeline to improve efficiency and shorten manufacturing time. In 2019, it took SpaceX engineers and technicians eight months to build the first Starship prototype. Half a year later, SpaceX was able to build a Starship prototype in just one month.

When SpaceX designs its rockets, it reuses parts as much as possible. The main hull of Starship is composed of dozens of welded stainless steel “vats” two meters high and nine meters in diameter. These stainless steel barrels are also the main body of the Super Heavy rocket. Musk said it was just a longer Starship fuel tank with more engines on the bottom: “When you build a Starship assembly line, you basically have an assembly line that produces Super Heavy as well.”

Use more small engines instead of developing bigger engines

Starship can become the most powerful rocket in human history not because its engine is stronger. The Raptor engine it uses has just reached the level of the RS-25 engine used on the US space shuttle in the 1970s, only 1/3 of the engine F-1 on the Saturn V.

But SpaceX uses a lot more. There are 6 Raptor engines on board Starship. There are 33 Raptor engines on a Super Heavy rocket. In traditional rocket design, using more engines is not optimal. Each engine is a steel behemoth, it is difficult to control one, and the difficulty will rise sharply if you want to control a group.

3e91690e8c9b4371196f5aa949480c8b.png

The 33 engines on the Starship booster. Source: SpaceX

A typical example is the Soviet Union’s N1 rocket. In order to surpass the Saturn V, Soviet engineers installed 30 engines on the N1, and the theoretical thrust reached the historical peak. But with so many engines, it’s impossible to fire them all at the same time, or better steer them. The N1 rocket was fired 4 times and blew up 4 times. The Soviet Union therefore shelved the moon landing plan.

In February 2018, when SpaceX launched the Falcon Heavy rocket with 27 engines for the first time, Musk said that SpaceX would not repeat the mistakes of the N1 rocket, they would do better.

Part of SpaceX’s confidence came from computer developments that the Soviet Union had not caught up with—particularly GPUs. In March 2015, SpaceX engineers demonstrated how they used GPU drivers for simulation testing when developing the Raptor engine.

72fe4760c9253e58c19b58641afb99fd.gif

SpaceX uses GPUs to simulate the operation of rocket engines. Image from NVIDIA GTC 2015.

SpaceX engineers will simulate what happens when the engine is running and look for potential problems. For example, when the fuel is not fully burned, the engine will keep shaking, causing the rocket to fly unstable. In severe cases, the entire rocket will not be able to carry the load or even disintegrate. The existing technology has no way to perform “nuclear magnetic resonance” on the engine while it is running. Using software to simulate the state of each part is the lowest cost and most suitable method.

Musk said that SpaceX uses more engines on the rocket similar to modern computer systems. Google or Amazon use a large number of small computers. Even if a computer fails, it will not affect the use. If they only use one or a few mainframe computers, if there is a downtime, the whole system will stop. “The same is true for rocket engines. Even if six engines fail on the Falcon Heavy rocket, it can still go to orbit.”

When the Starship lifted off this time, although 3 engines were not working, it did not affect its continued upward movement. The computer quickly adjusted the other engines to try to keep the flight stable.

Using more small engines can save SpaceX money. When sending larger rockets into space to carry more cargo or people, they don’t need to invest a lot of time, talent and money in developing engines with more powerful thrust and record-breaking power. SpaceX did not develop new engines for the Falcon Heavy rocket with more load, but directly used three Falcon 9s as boosters, a total of 27 engines.

Building more engines can also spread the cost of SpaceX, making the assembly line run more efficiently. In 2015, NASA allocated US$1.16 billion to allow Aerojet Rocketdyne, an American aviation and defense company, to restart the RS-25 engine manufacturing line to supply the engine for the SLS rocket in the return to the moon program. The annual production capacity is four units, each costing hundreds of millions USD – once and forfeit.

By November 2022, it will take less than a week for SpaceX to produce the same number of Raptor engines, each costing millions of dollars, and Musk hopes to reduce it to hundreds of thousands of dollars in the future. NASA Deputy Administrator Mark Kirasich (Mark Kirasich) said that SpaceX continues to iterate the Raptor engine, improving performance and thrust while reducing the number of parts and making the manufacturing and testing time shorter. “They can now build seven units a week. engine.”

In 2010, nearly 40 years after the end of the Apollo program, a small number of governments still controlled space activities, Russia was the country that launched the most rockets on earth, and NASA’s main force was the soon-to-retire space shuttle. Their center of gravity is no longer to go to farther places, but to deploy satellites around the earth. The reality is so far from the miracle that many people doubt whether humans have really been to the moon.

“In 1969, humans went to the moon. Then we had the space shuttle, which could only send people to low orbit. Then the space shuttle also retired.” In April 2017, Musk participated in the TED conference and said, “People have a mistake The concept: Technology will advance by itself. Technology will not automatically advance, and often even degrades. Only a lot of people work together to make it better.”

“TECH TUESDAY” series

In 1957, a man-made object entered space for the first time, orbiting the Earth for three weeks. Human beings can look up and see a small flash across the sky in the night, parallel to the stars in mythology.

Such feats have inspired joy across races and ideologies across the globe. But not in the triumphant joy we might have guessed, touched by human feats. According to the political philosopher Hannah Arendt (Hannah Arendt) observed back then, people’s mood is closer to a long-awaited relief that science has finally caught up with expectations, “Humanity is finally on the way out of the cage of the earth. took the first step.”

People are always rapidly adjusting their expectations of the world based on technological exploration. When a fantasy of a science fiction writer becomes a reality, it is often that technology has finally caught up with people’s expectations, or in Arendt’s words, “technology has realized and affirmed that people’s dreams are neither crazy nor empty.”

At times like today, a little more dreaming is better.

This is also the expectation of “LatePost” launching the TECH TUESDAY column. We hope to regularly report on new scientific research and technological developments outside of the business world that “Later” focuses on on a daily basis.

These may be about the progress of a cutting-edge research, the observation of a technology application, or a tribute to some outstanding technologies or even an era.

This column will record the various changes in the world from the perspective of science and technology. During this journey, I hope readers can join us in gaining a little understanding of the world.

This article is transferred from: https://www.latepost.com/news/dj_detail?id=1613
This site is only for collection, and the copyright belongs to the original author.