Application Technology

Space is becoming big business worldwide, which means launch rates must increase if they are not to be limited by launch availability. However, launches are still a risky affair and launch failures are not uncommon. For instance, in August a test was being carried out on Unst – an island in the far north of Scotland – by German company Rocket Factory Augsburg and the nine engine test vehicle exploded. 

Precise details of the cause will take a while to discover, but the fundamental problem is that rocket engines have to run close to mechanical and thermal failure to make them efficient. The slightest manufacturing or procedural error can lead to an explosion. Many will know about the Space Shuttle Challenger, which exploded in early 1986, killing seven astronauts. A leaking solid rocket booster was the cause. 

Re-entry is equally demanding – spacecraft re-entering Earth’s atmosphere do so at around 17,000mph, meaning the frictional heating is huge. Any flaw in the heat shield could also be disastrous. 

While most existing or planned launch services are targeting ’run of the mill’ spacecraft customers, Elon Musk’s SpaceX is looking much further ahead – to Mars, possibly via the Moon. The company made 96 successful launches last year of its smaller rockets, putting satellites into Earth’s orbit, but SpaceX’s Super Heavy launcher and Starship space vehicle are aiming much further into space.

From boyhood dreams to reality

Elon Musk has been enthralled with space since he was young, with his love of space stemming from science fiction and science fact. Today, as one of the world’s most successful people, he is about to launch his fifth Starship atop a Super Heavy booster. 

So long as no one is hurt or killed, Musk isn’t too worried about launch vehicle explosions. His philosophy is “let’s try it and see”. Failure is a great educator. He also has an unusual attitude to material selection and therefore fasteners. 

SpaceX’s Super Heavy launcher and Starship space vehicle are truly remarkable, but do not always use the latest materials and systems because of the drawbacks. They are basically made from steel, not aluminium alloy or composite materials, as you might expect. These were looked at, but the sheer size of the rocket made composites impractical – where can you get an oven or autoclave that big? 

Aluminium alloy is light but doesn’t tolerate heat very well. That’s why the Space Shuttle had a complex and expensive thermal protection system – with a major bugbear being the cost in its reusability. It was one of the reasons the Space Shuttle was nowhere near as cheap to operate as had been predicted. The spacecraft also lost tiles on launch and re-entry and they had to be replaced at significant cost. Every tile had a unique shape and had to be specially manufactured. Before each launch, each tile had to be pull tested to ensure it wouldn’t come off. 

So, SpaceX opted for steel and a simpler heat shield with common tiles. However, the use of steel meant many welds and SpaceX had to do lots of development to get the welding right. There’s an important lesson here, traditional materials can solve your problems, but sometimes they need advanced joining techniques – it can take a while to get them optimised. 

Improving image

The first Starship looked very rough. Today, it is much smoother with less apparent welds. Will it soon be completely smooth and shiny as in SpaceX promotional pictures and videos? Image is important – even for a space launch company.

Originally, Starship was to be made from carbon fibre reinforced plastic (CFRP) with production beginning in California. This seemed logical since it’s extremely light and strong and most of the workforce was in California. It was therefore surprising when a stainless steel Starship ended up being made in Texas, but there were a number of reasons as to why. 

CFRP begins to break down at about 200°C. It would therefore require a very good heat shield to survive the 1,600°C encountered on multiple re-entries. Stainless steel is heavier, but it can stand up to 850°C, so a thinner heat shield is sufficient. Also, CFRP costs about US$150 per kg, whereas stainless steel costs a mere US$3 per kg and that’s a huge difference. 

Initially, Starship was made out of rings of 301 stainless steel that was 4.5mm thick. The company that made it built water towers – with no experience of building rockets. The welding standards were poor. To begin with, flux-core welding was used, in which current is passed through a metal wire forming an arc of about 7,000°C. The tip of the wire melts and falls into the molten steel, filling any cracks or air bubbles. A material around the wire burns and releases a protective gas around the weld – preventing it from oxidising. 

However, at this time SpaceX didn’t have a suitable factory and most of the welding was done outside. This resulted in the wind creating poor welds because it blew the protective gas away. The welds were heavily corroded and had cracks and rough edges on the surface. To manage this SpaceX started grinding the welds down, which perhaps surprisingly strengthened them because cracks on the surface can propagate through. 

Unfortunately, Starship Mk1 exploded when one of the welds failed, which led to more improvements being made. Thinner grades of 304L stainless were used, which needed much less welding. Also, it was much more resistant to corroding. The company also went to TIP TIG welding, allowing a tighter, deeper weld. This also warped the surrounding metal less. SpaceX also got robotic welding machines, similar to the ones in Tesla factories. Stringers added inside the rocket also stopped it buckling under its own weight. 

SpaceX’s steel is cold rolled, which strengthens it by stretching the grains. It also makes it harder. However, welding causes the material to soften. So, SpaceX got a giant planishing machine, which hammers the welds and compresses them till they match the hardness of the surrounding metal. It also improves their look. The company is also now developing its use of laser welding. 

Heat shield woes

Even though stainless steel has good heat resistance, SpaceX’s Starship still needs a thermal protection system and this is made up of a thermal blanket with tiles on top, similar in concept to the ones used on the Space Shuttle, but with far fewer shapes. The problem has been keeping them on. Some are fastened with metal studs and others are bonded on – with SpaceX also experimenting with the adhesives for the tiles.

The tiles are very similar in chemical composition to the ones used on the Space Shuttle (essentially a porous silicon material), but it is not clear at this stage why some are coming off. One theory is the booster, which is filled with cryogenic propellant that causes it to contract. On launch, it produces lots of vibration, and noise, and heats rapidly. All these things tend to dislodge or crack the tiles. Musk is adding an additional ablative material under all of Starship’s existing heat shield tiles. This will provide extra protection if a tile breaks or falls off.

During the last flight, Starship’s upper stage managed to make it through Earth re-entry despite a weak point between a flap and the main body that nearly destroyed the flap. Even so, Starship managed to complete its landing burn anyway with the flap still actuating.

Musk talked about a change to the entire heat shield of Starship. During a livestream, he said: “We’re going to replace the whole heat shield on the ship. So, the new heat shield tile is about twice as strong as the ones that were on the last flight. We want to put an ablative secondary structure, basically ablative protection behind the tiles, so that if the tiles crack or come loose it doesn’t cook the rocket.”

He continued: “We ran two experiments with tiles on this flight, which people may have noticed because there were missing tiles. They were intentionally missing because we were testing the secondary heat shield material, which is like a silicone felt ablative.” 

In pictures taken before the flight, these missing tiles were obvious toward the bottom of the ship. What wasn’t so clear was the addition of the ablative material, which must have performed well in those specific areas considering SpaceX has now added an entire layer behind the tiles of the next ship. 

Musk went on to clarify: “It’s not good for reuse but it’s good for saving your butt if a tile cracks or falls off. It’s very tricky to put these tiles on and have them work well because the tiles are ceramics; they are like a coffee cup or a dinner plate. So, you have a whole bunch of dinner plates on a rocket that is shaking. It’s shrinking cryogenically with the propellant and then expanding under pressure and then the tiles are expanding when they get hot. So, there’s a lot of expansion and contraction happening while trying to keep all these brittle tiles from cracking or breaking off.” 

He went on to talk about the challenges of making a reusable orbital heat shield, citing the Space Shuttle taking months of work between flights to fix and repair its heat shield. What’s clear from these comments is that they are still struggling a bit with the main heat shield tiles themselves, as far as keeping all of them secure and in one piece throughout the entire launch process. However, the new ablative material, in combination with the rocket’s steel construction, is expected to provide a significant safety barrier if a few tiles fail.

The goal for Starship is rapid reuse; this obviously would not be possible if after each landing the rocket needs repairs to the underlying ablative material, as well as lost tiles replacing. The other specific fix had to do with the seal where plasma initially broke through and damaged the flap. In regard to that change, Musk initially tweeted that they would have this nailed for the next flight and that it didn’t seem to be a very significant alteration. 

He said again that they still want to catch the booster on the next flight. Specifically, he was quoted saying: “Unless something comes up that we think is problematic, we will try to bring the booster back and catch it with the giant Mechazilla arms (known as chop sticks).”

Another detail from this livestream, combined with other tweets and comments, suggests that only the one flap was damaged. Whenever mentioning the damage, it was always singular rather than plural. If that is the case and only the flap shown on camera was damaged, it would be an even better sign as the others held up even with the current heat shield system. With heat shields, a tiny weakness can easily become a disaster for the entire vehicle.

Based on everything said, we should expect to see SpaceX get very busy in the coming weeks in preparation for another launch. The launch pad seems to be in great shape, including the water deluge and steel plate protection. This system has now been used for three launches and seems to be holding up nearly perfectly.

In a post-launch report, SpaceX said: “Starship’s fourth flight test launched with ambitious goals, attempting to go farther than any previous test before and demonstrate capabilities central to the return and reuse of Starship and Super Heavy. The payload for this test was the data.” 

Musk made it clear that they got a lot of incredible data from Flight 4, but he pointed out that both the soft landing of the booster and upper stage was about a 20% lucky event, based on their initial expectations. With this came even more data they can apply to the next mission to try even more ambitious things. 

Once the catch process is attempted, depending on how that goes, it could become somewhat of a regular occurrence. From there we can hope to see the ship actually returning for an attempted touchdown on land. The faster Starship becomes operational the better. 

So, what are the main lessons here? Explore and ignore the naysayers. You don’t have to always use the latest technology. Processes have to offer value for money. Time served materials with new processes can be the best. Also, don’t be scared of trying something that fails. The opposite of success is not failure; the opposite of success is not trying. Failure is part of success and nearly all major successes had lots of failures on the way. 

With Super Heavy and Starship, the production welding and the thermal blanket and tile fastening system, whether it be with metal studs or adhesive, are crucial to the vehicle’s survival and re-use. If the fastening systems fail in many places, the vehicle will likely be lost. A few tile losses can be tolerated, but a major weld failure probably not. 

With SpaceX’s ambition to launch astronauts on Starship, the chances of a major failure must be as low as possible. The company’s ‘Try it and see’ attitude is producing a huge amount of data, hopefully meaning that any failures happen before astronauts are flown. The company’s attitude is very much ‘Test it in the real conditions’ rather than any simulation. 

SpaceX’s approach is the opposite of the careful, plodding, conservative development of other launch vehicles and it seems to work. Perhaps competitors will now follow suit.