Starship Update

Podcast Transcript

One of the biggest advances in space flight right now is happening with the development of Starship.

Starship is the largest rocket ever flown, but it is fundamentally different than all other rockets that have come before it. Its intent is to be fully reusable.

This has the potential to completely revolutionize spaceflight and dramatically reduce the cost of bringing cargo to orbit.

It’s not there yet, but it’s getting close.

Learn more about Starship and its latest advancements on this episode of Everything Everywhere Daily.

I did my first episode on Starship back in February of 2022, over three and a half years ago. I did a redux episode in January of 2024, where I gave a brief update, but it is necessary for a more comprehensive update on what is happening because there has been a lot of activity.

Before I get into the latest developments, let me give a recap of the problem that Starship is trying to solve.

Ever since the launch of Sputnik in 1957, launching anything into space has been difficult and expensive, with an emphasis on the expensive.

The reason why space has been so expensive is that, until recently, all rockets have been disposable. You build an entire rocket, launch it, and destroy it in the process, with most of it ending up either at the bottom of the ocean or burning up in the atmosphere

The costs for launching 1 kilogram, or 2.2 pounds, into low Earth orbit were about $50,000 on average in the 1960s, adjusted for inflation.

The high cost of launching anything into orbit was the main impetus behind the Space Shuttle program. The idea was that having a mostly reusable vehicle would reduce costs dramatically.

If you remember back to my episode on the Space Shuttle, that didn’t happen. Initially, the cost to orbit actually went up with the Space Shuttle. Because of the complexity and fragility of the shuttle, it never came close to achieving its goals for the number of flights.

During the entire life of the shuttle program, the cost per kilogram to low Earth orbit averaged about $54,000.

Costs decreased after the shuttle program simply because disposable rockets became cheaper, but the real groundbreaking drop in price came with the development of the Falcon 9 rocket by SpaceX.

The Falcon 9 was a breakthrough because the entire first stage of the rocket was reusable. Instead of landing horizontally like an airplane, the rocket landed vertically. It was an incredibly challenging engineering problem, but SpaceX managed not only to get it to work, but has now done so 548 times out of 551 attempts as of the recording of this episode.

That is a 99.5% success rate.

The Falcon 9 booster with ID number B1067 has been launched a record 27 times.

With the Falcon 9, the cost per kilogram to orbit plummeted to about $2500.

With the Falcon 9 Heavy variant, the cost can drop to $1,500.

This is an incredible reduction in price, but there is still a lot of room for improvement.

The second stage of the Falcon 9 still can’t be recovered, and the size of the rocket limits the amount that can be launched at one time, limiting possible efficiencies.

Starship is the ultimate goal of SpaceX. It is a fully reusable rocket, and it is also the largest rocket to have ever flown. It is larger than the Saturn V from the Apollo Program.

The goal is initially to reduce the cost per kilogram to orbit down to about $100. Depending on how many times they can reuse the components, it is possible to get that number down to $10 to $20.

In addition to being fully reusable, there are a host of additional innovations that are being implemented.

One big one is the number of engines being used. The Saturn V used five enormous F1 rocket engines on its first stage. Starship uses 33 Raptor engines in its first stage.

Using more, smaller engines is challenging. The Soviets tried doing this with their Saturn V competitor called the N1, which had 30 engines. They never got it to fly successfully.

Having more engines has benefits, however. It is cheaper to make a smaller engine. If you lose an engine, it is catastrophic. They are easier to swap out and refurbish without resulting in downtime.

The same Raptor engines are also used on the orbital second stage as well, meaning you don’t need multiple types of rocket engines.

They have also created a unique launch tower especially designed for Starship.

The launch tower for the Starship is known as “Mechazilla.” Not only does it support vehicle stacking, lifting the giant first-stage booster and then the second-stage ship onto the pad, it also houses giant mechanical arms designed to catch the returning first stage after launch, rather than relying on landing legs and wheels.

This approach reduces the mass and complexity of the booster, enables very fast reuse, and allows the rocket to be caught and immediately returned to the pad for re-launch, representing a fundamental shift in how launch hardware is recovered.

Compare this to what NASA did for years, where they had to assemble everything in an enormous building and then roll every rocket out to the launch pad.

So, since I did the first Starship episode, SpaceX has had 11 test launches of Starship. Perhaps you’ve seen them on YouTube, where each of them is usually live-streamed on multiple space-themed channels such as Everyday Astronaut and NASASpaceFlight.

The test flights are exactly that. They are designed to test various components and systems, then take the data they get from those tests and try to implement that data into future launches.

Many people have misunderstood the SpaceX approach because they are used to how NASA always did things.

The best analogy to what SpaceX is doing is that they are conducting their engineering similarly to how software is often written. A software program will often be released in alpha or beta versions to look for bugs and issues that need to be fixed before releasing a final product.

SpaceX is trying to iterate rapidly. Launch, find out what went wrong, and then do it again as soon as you can review the data and implement changes.

Of the eleven test flights of Starship that have been conducted so far, six can loosely be considered successful, and five have been considered, quote unquote, failures.

I say “quote unquote” because even the failures have been learning experiences.

The failures basically mean that the rocket blew up at some point in the flight, unexpectedly, or that the flight objectives were not achieved.

The first full test of the full stack took place on April 20, 2023. This set the record for the largest rocket launch, which was set by the Soviet N1. While the launch was successful, there were problems just a few minutes in, and the autodestruct system was activated.

Test 2 came almost half a year later, on November 18. This successfully achieved the separation of the two stages, but both stages were ultimately destroyed.

Test 3 was on March 14, 2024. It went better. The second stage achieved orbital velocity but broke up over the Indian Ocean.

Test 4 was the fastest turnaround yet and took place on June 6. This was the best test yet. They simulated a near-pad “virtual tower” booster landing above the Gulf and had the ship survive peak heating during reentry and reach a controlled splashdown corridor in the Indian Ocean.

Test 5 was a big one. On October 13, they caught Booster with the chopsticks, and the ship completed ascent and splashdown testing.

Test 6 was barely a month later, on November 19. It was similar to test 5, except the ship completed a Raptor engine ignition in space. Another landing on Mechazilla was scrubbed at the last minute.

This completed their block one era. These were the first iterations of both rocket stages, and the lessons learned in this phase were then put into a new design for Block Two.

The new block two version of Starship featured enlarged propellant tanks that increase total capacity and thrust duration, re-engineered aerodynamic flaps with relocated hinges to reduce heating during reentry, and a more robust heat-shield system that uses pinned tiles, improved insulation, and tighter seams for better durability.

The interstage and plumbing layouts were simplified for lighter mass and easier maintenance, while avionics and venting systems were modernized for greater reliability. Together, these changes make Block 2 heavier, taller, and far more capable.

All of the block two flights took place in 2025.

Test 7 was on January 16. The goal was to just repeat tests 5 and 6 with the new model, but the ship suffered a propellant leak that cost attitude control, and the flight was terminated.

Test 8 was on March 6. The booster phase went well, and it was able to land on Mechazilla. However, the second stage lost multiple engines and then telemetry, leading to a breakup before the planned objectives were complete.

Test 9 was on May 27. This didn’t go so well. The ship reached its planned velocity, then a propellant system leak and loss of attitude control curtailed the mission. The booster also broke up after starting its landing burn.

Test 10 was on August 26. For the first time, Starship managed to deploy dummy satellites. However, an engine hardware failure resulted in the ship not being able to do a soft landing.

Test 11, the most recent test so far, took place on October 13. This was the last of the block 2 tests. It was an almost perfect test, which achieved everything that was set out, including a soft landing of both the booster and the ship.

What you can see over the course of the tests is the iterative improvement in the rockets. Sometimes it’s one step back and two steps forward, but things are moving forward, which is a lot considering this is both the largest and most complicated rocket to ever fly.

The next tests, which should begin in early 2026, will be the start of the Block 3 Starship builds.

One of the biggest changes will be the adoption of the version 3 Raptor engines. These are actually a remarkable bit of engineering. If you look at a photo of the version 1 versus the version 3 engines, they are night and day.

The version 1 engines look incredibly complicated with tons of external pipes and tubes. The version 3 engines are smaller, sleeker, and more powerful at the same time.

Most of the wires and pipes have been integrated into the engine itself, and much of the engine is now being built via 3D printing, which eliminates the need for many bolts and welds.

It also simplifies the aft end of the rocket by eliminating most external engine shielding, which isn’t necessary with the new design.

The booster gets a redesigned forward dome and vented interstage, plus three larger grid fins double as tower catch points; the ship adopts a revised heat-shield tile design and dedicated on-orbit refueling hardware, including a docking interface and a new quick-disconnect.

In addition to testing new hardware, the plan is to test new techniques. The biggest of which is to conduct the first fuel transfer in orbit, which has never been done before. This will also result in the first full orbital missions.

There is also a chance that specially designed ships, which are not intended to return to Earth, might be sent to the Moon or Mars in 2026. These would be tests to see if they can land remotely on another body and take off again.

A Mars mission might just be a Hail Mary pass because Mars happens to have a launch window this year. If they don’t try now, they’d have to wait another two years.

That being said, a moon mission test might not take place until 2027. Nothing has been made public yet.

There is still a lot of work to be done on Starship, and I’m sure I’ll be doing yet another follow-up episode at some point in the future, but things are most definitely happening and are going in the right direction.

If you haven’t watched any of the launches, I recommend you do so, because they are really fun. There are probably over a dozen YouTube channels that provide live coverage of the launches and provide excellent commentary on what is happening.

This is one of the most exciting times in spaceflight since the start of the Shuttle Program back in the early 1980s.