Starship SN9 | 10 kilometer flight

Launch Window/Lift Off Time
(Subject to change)
NET February 2, 2021
Mission Name
10 km flight test
Launch Provider
(What rocket company is launching it?)
(Who’s paying for this?)
Starship SN9
Launch Location
Boca Chica Launch Site, Boca Chica, Texas
Payload mass
There is no payload on this flight
Will they be attempting to recover the first stage?
Where will the first stage land?
They will attempt to propulsively land SN9 on a landing pad in Boca Chica
Will they be attempting to recover the fairings?
The fairing (or more accurately, the nosecone) is integrated into the vehicle and will be recovered with it
How’s the weather looking?
This will be the:
1st flight of SN9
– 2nd high-altitude flight of a Starship prototype
– 4th flight of a Starship prototype (6th if you count Starhopper
’s two flights; 8th if you count Starhopper’s two short tethered hops as well)
– 2nd flight of a prototype with a nosecone and aerodynamic control surfaces

– 2nd flight of a Starship prototype with three engines
Where to watch
Official livestream

Tim Dodd, the Everyday Astronaut, will be streaming.

Live Updates on the lead-up to SN9’s flight

What does all this mean?

SpaceX will be conducting a 10 km test flight of their SN9 prototype vehicle. This flight will be a repeat of the largely successful SN8 test flight, with some minor flight profile changes and a lower apogee.

SN9 will lift off from Boca Chica, Texas under the power of its three Raptor engines. The Raptor engines will shut off one by one during ascent to decrease loads and acceleration on the vehicle. At apogee, the single Raptor engine still burning will initiate the vertical-to-horizontal flip before shutting off. SN9 will make use of its body flaps to keep it stable during descent. Prior to landing, SN9 will ignite two of its Raptor engines to reorientate itself back into a vertical position and perform a propulsive landing.

What is Starship?

Starship/Super Heavy is SpaceX’s next-generation super heavy-lift reusable launch vehicle. The first stage, known as Super Heavy, will produce 71.2 MN of thrust, while the Saturn V, the most powerful rocket ever flown, produced 34 MN of thrust. This will make it the most powerful rocket ever made. When it’s operational, Starship will liftoff from either Boca Chica, Texas; Cape Canaveral, Florida; or an offshore launch platform.

Starship is SpaceX’s plan to colonize Mars and dramatically bring down the cost of space travel. With an estimated eventual cost of $2 million per flight, Starship will be one of the cheapest orbital rockets ever, especially on a cost-per-kilogram basis. Starship will be able to carry 100 tonnes to just about anywhere, due to orbital refuelling.

Starship variants

There will be several Starship variants including a cargo variant with a clam shell-like opening to deploy satellites, a crewed variant with life support systems and crew habitation areas in place of a cargo bay, and a lunar variant without flaps and with additional thrusters.

Since Starship is the second stage of the rocket and not just a spaceship, it will have to use most of its fuel to insert itself into orbit. To enable Starship to carry out deep-space missions with 100-tonne payloads, SpaceX will also have a tanker Starship variant. This variant will be similar in appearance to the cargo version, but will have additional fuel tanks instead of a payload bay. This Starship will dock by the engine bay with another Starship carrying either crew or cargo in low Earth orbit (LEO) and refuel it. It may take several rendezvous and refuellings in LEO before a Starship has enough fuel to continue its journey deeper into space.

Starship variants (Credit: Everyday Astronaut / Caspar Stanley)

The future of Starship

Starship is designed to revolutionize travel not only in space, but on Earth as well. If Starship’s goals for airplane-like safety and reliability pan out, it could be used for suborbital point-to-point Earth travel. With a relatively low price tag, a fairly average person could be anywhere in the world in less than an hour.

After stage separation, the Super Heavy booster will perform a boost-back burn and a re-entry burn to slow itself down enough to return to launch site. The booster will perform a fourth and final burn to land either on or near the launch pad.

After the completion of its mission, Starship will perform a deorbit burn then reorientate to re-enter heat shield-first. Reentry will bleed off most of the vehicle’s velocity. Just prior to landing, Starship will reorientate itself again to perform a propulsive vertical landing. Starship and Super Heavy will then be restacked, refueled, then launched again.


Raptor is SpaceX’s newest rocket engine. It’s the only full-flow staged combustion cycle engine to have ever flown. It’s able to produce about 2,300 kN (500,000 lbs) of thrust, making it twice as powerful as SpaceX’s workhorse Merlin engine.

The fact that Raptor is a full-flow staged combustion cycle (FFSCC) engine makes it unique among rocket engines. In a closed cycle rocket engine (which Raptor is), the fuel used to spin up the turbines is directed into the combustion chamber rather than being ejected out of a preburner. This means all fuel fed into the engine is combusted, resulting in a more efficient engine. Additionally, all of the fuel is burnt in the preburners and sent to the combustion chamber as gas, which allows Raptor to achieve higher pressures, and in turn, higher thrust and specific impulse.

History of Starship

Interplanetary Transport System

The history of Starship goes back to at least 2012 when it was known as Mars Colonial Transporter. At that time, it wasn’t much more than a few murmurings from Elon, but more detailed plans were unveiled in 2016 at the International Astronautical Congress. A two-stage carbon-fiber vehicle was announced. The vehicle, now dubbed Interplanetary Transport System (ITS), would be 122 meters (400 ft) tall and 12 meters in diameter. In an expendable configuration, the vehicle could carry over 500 tonnes to Low Earth Orbit (LEO). In a reusable configuration, it could carry 300 tonnes to LEO, more than double that of the Saturn V.

ITS and Saturn V
ITS and Saturn V (Credit: SpaceX)

Big Falcon Rocket

At the 2017 International Astronautics Congress, a major redesign was announced. The name was changed to Big Falcon Rocket (BFR). It was downsized slightly, to only 9 meters in diameter. Small delta wings were added to the vehicle to control it and induce drag during re-entry. The BFR was a smaller and more feasible design, which was intended to launch to Mars as a cargo variant in 2022, with a crewed flight following two years later. Construction of a prototype vehicle started in early 2018 at the Port of Los Angeles but was later scrapped.


In September of 2018, SpaceX announced another major redesign. The vehicle now featured three rear fins, that doubled as landing legs, and two forward canards for control during atmospheric re-entry. At the same event, it was also announced that Japanese billionaire Yusaku Maezawa was partially funding Starship development and would be one of the passengers during Starship’s first crewed trip around the moon. He, along with several different types of artists, is planned to take a flight around the moon in Starship as part of the #dearMoon project, with the goal of inspiring the artists to create works that inspire people back on Earth. A few months after this event the vehicle was renamed to Starship/Superheavy.

The big switch

In December 2018, Elon Musk announced that Starship had switched from being built out of carbon-fiber to being built out of stainless steel. He explained that the particular alloy they were using was significantly cheaper than carbon-fiber and got 50 percent stronger at cryogenic temperatures. The material is also tough at high temperatures, meaning it needs less heat shielding compared to what the previous carbon-fiber design would have required. The parts of the vehicle not exposed to the most intense heat would be cooled by liquid fuel “sweating” through tiny pores in the steel.


Around the time of the announcement of the new stainless steel design, SpaceX began building its first prototype: Starhopper. The vehicle was built outdoors in Boca Chica, Texas, a small village near the Mexican border. SpaceX acquired property in Boca Chica in 2014, but until 2018 the area sat largely dormant with little activity. Occasional dirt deliveries were the most exciting thing happening in the four-year interim period. Late spring of 2019 saw the completion of Starhopper. Initially, Starhopper was built with a nosecone, but a windstorm in January caused it to fall off the vehicle and collapse. Since the nosecone was not necessary for flight, SpaceX decided to forgo it and launch without.

Starhopper hops!

On July 25, 2019, Starhopper completed its first untethered hop, successfully reaching an altitude of 20 meters, before performing a soft landing. A month later, on August 27, Starhopper completed a 150-meter hop. After this flight, Starhopper was retired and SpaceX started going full-steam-ahead with MK1.

Starhopper’s 150-meter hop (Credit: Everyday Astronaut)

Starship Mk1 and Mk2

While testing Starhopper, SpaceX also began constructing its first of two full-scale prototype vehicles: Mk1 in Boca Chica and Mk2 in Cocoa, Florida. Elon Musk announced on Twitter that the teams at each facility were in a race to orbit. Each team constructed the vehicles using their own techniques, but they were required to share anything they learned along the way. Because SpaceX had already started constructing their first two full-scale Starship prototypes as Starhopper was undergoing testing, for the first time 2019’s Starship presentation took place in front of an actual Starship mock-up. Mk1 was rolled out to the launch site on October 30, 2019.

Mk1 testing and Mk2’s fate

Mk1’s testing began with a cryogenic pressure test on November 30, 2019. This test was meant to make sure Mk1 could hold up to the high pressure and low temperatures that would occur during flight. Due to bad welds, the vehicle couldn’t handle the pressure and exploded in a cloud of super-chilled nitrogen. The loss of Mk1 wasn’t a major one, as SpaceX wasn’t expecting to ever fly the vehicle.

MK-1 failure
Mk1 over-pressurization event (Credit: BocaChicaGal for NASASpaceFlight)

Meanwhile, work on Mk2 in Cocoa, Florida slowed to a halt. The site was largely abandoned in favor of focusing on progress in Boca Chica. Mk2 sat completely dormant for months until SpaceX started deconstructing it in July of 2020.

New Rings

Mk1’s construction process and failure revealed issues that would prevent a Starship of its construction method from ever flying. SpaceX radically changed its construction process, not only to decrease the number of welds but also to decrease the mass. Mk1 was made up of large steel plates welded into rings and stacked on top of each other. Not only did that create many weak points, but it also made building the vehicle laborious and time-consuming. With the purchase of an IMCAR machine, SpaceX began making each ring out of a single sheet of rolled steel. IMCAR is an Italian company that specializes in making machines that roll steel into circular barrels.

Starship rings
IMCAR machine (Credit: Italian Food Tech)

SN1, SN2, SN3, and the early test tanks

In the early part of 2020, we saw the construction and completion of two scaled-down Starship test tanks. These two tanks used different welding techniques and had improved bulkhead designs. Both were tested to destruction at the launch site.

Following the successful destruction of the test tanks, SpaceX continued with SN1’s construction. Their first full-scale prototype with improved rings, bulkheads, and welding techniques was completed in February. Cryogenic testing occurred on February 28, 2020.

SN1 testing

SN1 failed during its first cryogenic test when its thrust puck failed. The thrust puck is the bottom bulkhead that doubles as an attachment point for Raptor engines. The thrust puck must be able to handle the force of the thrust as well as the pressure from the liquid oxygen tank. The failure sent the vehicle several meters into the air.

Thrust puck redesign

SpaceX temporarily refocused their efforts on the thrust puck. They completely redesigned it to be able to withstand the 7 bar necessary for flight. They built a scaled-down prototype, SN2, which was similar in size to the pre-SN1 test tanks, to test the new design. SN2 passed cryogenic testing with flying colors and was retired to a concrete stand near SpaceX’s scrapyard.

Starship SN3

SN3 was the next prototype and sported the newly redesigned thrust puck. It was smoother than SN1 and had many improvements. Unfortunately, a testing configuration error caused the vehicle to lose pressure in its lower tank and collapse during a cryogenic test. It wasn’t the fault of the vehicle this time, but actually the fault of the ground support equipment (GSE). The GSE was unable to keep the two tanks in the vehicle at similar pressures. The lower tank crumpled under the weight of the upper tank.

SN3 failure (Credit: BocaChicaGal for NASASpaceFlight)

Starship SN4 and first static fires

After the SN3 failure, SpaceX quickly churned out the next prototype: SN4. It was again noticeably smoother than the previous prototypes and reused the lower part of SN3’s undamaged skirt section. On April 26, 2020, SN4 passed cryogenic testing, becoming the first full-sized Starship prototype to do so. It reached a pressure of 4.9 bar, which Elon Musk said was “a softball tbh, but that’s enough to fly!” A Raptor engine was installed the next day, and SN4 static fired for the first time on May 5, then again only a day later on May 6. The first static fire fed fuel from the main tanks while the second one fed fuel from the main oxygen tank and the smaller methane header tank. Following the two successful static fires, the Raptor engine was uninstalled.

More SN4 testing

SN4 underwent another cryogenic test, but this time reached a pressure of 7.5 bar, enough for an orbital flight. Following that, a different Raptor engine, Raptor SN20, was installed. That engine static fired for the first time on May 19. During the static fire, a methane line was knocked loose from the vehicle, and slightly scorched it. SpaceX was unable to safely depressurize the vehicle, so no one was allowed to be near SN4 for two days while its fuel slowly boiled off. Fortunately, the repairs were minor and the vehicle static fired again on May 29.

SN4 failure

The static fire was successful, but the vehicle started leaking oxygen and methane, due to a GSE pipe becoming disconnected. Moments after the static fire, the vehicle exploded.

SN-4 failure (Credit: Everyday Astronaut with @SpacePadreIsle on Twitter)

The damage was greater than any Starship failure before. The explosion punctured a hole in a nearby water tank and damaged the launch stand beyond repair. Fortunately, SpaceX was already constructing a second launch stand, so replacing the old one only took a few days.

The cause of SN4’s failure

It was later found out that the quick disconnect system was responsible for the failure. The quick disconnect system disconnects the vehicle from ground support equipment prior to launch. A fuel line was knocked loose and created a large cloud around SN4. A hotspot in the engine bay ignited the cloud and detonated the vehicle. The quick disconnect test was expected to be an easy test but ended up causing a catastrophic failure.

Starship SN7 and SN7.1

SN7 followed SN4. It was made of an entirely different steel alloy. The previous Starship prototypes used 301 stainless steel, while SN7 used 304L stainless steel. 304L steel is stronger under cryogenic temperatures. During its first cryogenic pressure test, intended to be to destruction, the vehicle didn’t pop, but rather, sprung a relatively small leak at 7.61 bar. SpaceX reinforced the failure point and conducted another pressure test a few days later. This time, the vehicle failed at the bottom bulkhead’s weld line and went several meters into the air and to the side, taking its test stand with it. The pressure SN7 reached during its second test is still unknown but was presumably an acceptable pressure. SN7.1 was tested a couple of months later. SN7.1, like SN2, tested a new thrust puck design. After several aborted attempts, SN7.1 was tested to destruction on September 23.

Starship SN5 and SN6

SN5, after waiting in the wings for several weeks, rolled out to the pad on June 24. On July 1, SN5 passed cryogenic testing and static fired on July 30. Five days later, on August 4, SN5 became the first Starship prototype to successfully hop.

SN5 in flight (Credit: Elon Musk on Twitter)

Following SN5’s successful 150-meter hop, Elon Musk revealed on Twitter that SpaceX would “… do several short hops to smooth out launch process…” About a month later, on September 3, SN6 became the second Starship prototype to successfully hop to 150 meters.



On December 9, SN8 lifted off from its launch mount and began the first high-altitude test under the power of three Raptor engines. At around 1 minute and 40 seconds into flight, SN8 intentionally shut off one of its engines to decrease acceleration and loads on the vehicle. At around 3 minutes and 10 seconds into flight a second Raptor engine was intentionally shut off for the same reason. Raptor SN42 continued to burn up until T+4 minutes and 40 seconds. Just prior to its shutoff, it gimbaled to initiate the flip maneuver.

Descent and landing

SN8 showed impressive stability and control with its flaps during descent. The vehicle remained in a solid horizontal configuration throughout the entire skydive phase of its flight. As SN8 approached the ground, two of its Raptor engines ignited one by one to initiate the final flip maneuver in tandem with its body flaps. Back in a vertical orientation, SN8’s landing burn continued. However, in the last seconds of flight, the methane header tank lost pressure. The vehicle was now unable to produce enough thrust for a soft landing and began accelerating. One of the engines shut off while the other one continued to burn, creating a green plume because the copper components of the engine itself were combusting due to the absence of methane.

SN8 was unable to slow itself down enough and impacted the ground at a high velocity, resulting in a rapid unscheduled disassembly (RUD). Though the vehicle was reduced to a pile of rubble, the flight was considered a smashing success. Almost everything went flawlessly on the first high-altitude flight, including the daring bellyflop maneuver that many people were concerned about.

SN8 successfully demonstrated that the horizontal-to-vertical landing flip was not only possible, but able to be achieved during the first attempt. It also proved that the body flaps were able to effectively control the vehicle and keep it stable.

SN8 bellyflop (Credit: Everyday Astronaut)


SN9 will perform a 10 km test flight – the second high altitude flight of a Starship prototype. The flight will be similar to SN8’s, just with a lower apogee. SpaceX will pressurize SN9’s header tanks with helium for this flight, which should prevent them from losing pressure like the methane header tank did during SN8’s flight.

SN9 tipped over in the high bay the day after SN8’s flight, causing a delay to its planned rollout while SpaceX inspected and replaced damaged parts – namely, one of the forward body flaps that was crushed between the vehicle and the wall of the high bay. Starship SN9 finally rolled out just before Christmas, with two of its three engines already installed – its third engine was installed at the launch site.

SN9 conducted two cryo tests, followed by a static fire. Several days later, SN9 conducted three static fire tests on the same day. Two of the engines were damaged following the trifecta of static fires. They were swapped out and SN9 successfully static fired for a fifth time.

  1. Would it not make more sense to build a fuel station in orbit that the cargo/crew starships connect once and fully fuel themselves? It should keep the fuel cold enough even in the sun as it could have condensors / chillersto maximize fuel load. If built correctly you could have multiple connection points to allow multiple starships to dock for filling or refill

    1. Instead of building a ‘Fuel Station’ in orbit why not just use a specially equipped ‘Tanker Starship’ instead?

    2. Ulasan mengagumkan tentang sejarah Starship, sangat menarik dan informatif! Terima kasih

  2. Takes me back to my Estes model rocketry days and getting my Big Bertha to land on its tail fins.

  3. Молодцы ребята ! Не в теории годами сидят, а на практике (сразу!) теорию проверяют. Спасибо Илону !

      1. Great article – Typo on Re-enter currently “renter” in this line
        “will perform a deorbit burn then reorientate to renter heat”

      2. Another typo, should be “of” not “if”
        SN9 will make use if its body flaps to keep it stable during descent

  4. It will be 200 years before any human being has a moony haksent the word is spelt in plain English with the brain and tongue not the eye and hand. The only thing the space race has is anti gravity toilets. You need anti gravity engines
    To get you anywhere in the space race it e will be 1000 years before the human race will understand the working of anti gravity engines. The next best thing is electromagnet turbo fans which have a limitless range and only make a noise like a large bee . And would get you to the moon within 12 hours.p

    1. What, exactly, would those ‘electormagnet(ic) turbo fans’ be using to produce thrust between the Earth and the moon? You are aware that there is no air in space, are you not? Turbofan engines work inside an atmosphere, but lacking anything to exert force against, they would be so much spinning nothing. One other small factor, that ‘large bee’ sound also relies upon an atmosphere to propagate sound waves.

      I cannot even comment on the majority of your comment, as it just plain doesn’t make any sense.

      Unless, of course, your comment is meant as humor.

  5. Mr. Dodd, a clear, informative summary of the Starship program! It really does a nice job of taking all the bits and pieces of SpaceX news and laying them out clearly. Much easier now to see Starship’s progress. Thanks!

  6. Yeey. Thumbs up (and fingers crossed) it all goes 100% nominal 😉 on this flight!!

    Bok from Zagreb

  7. I am not sure what definition you are using, but both the RD-180 and the SSME are full flow staged combustion engines, the first using Kerosene as the fuel and the latter Hydrogen. The Soviets also developed the NK-33/34 using Kerosene and the RD-0120 using hydrogen. Raptor is the fifth such engine, and the first to use Methane as the fuel. There are minor differences in how each engine pre-burner operates, but all exhaust all their combustion products through the main nozzle, generating thrust.

    1. The definition from the Wikipedia article you linked is typical: “Full-flow staged combustion (FFSC) is a twin-shaft staged combustion cycle that uses both oxidizer-rich and fuel-rich preburners.”

      RD-180 is not full flow, it only has one preburner which is oxygen rich.

      I believe both preburners on the RS-25 (ssme) are fuel rich, so it’s doesn’t quite count as full flow.

      I don’t really know anything about the other engines you listed.

  8. Hello I am from Madeira island, Portugal, and a fan of SpaceX. All will be different after Starship on space transports, is a wonderfull project. Keep walking and good luck. Health to everybody and thanks to Mr. Elon Musk.

  9. The very top of the article says “NET January 6, 2021.” I am new to following this. What does NET mean in this context?

  10. The SN9 Prelaunch Preview states an expected 12.5KM flight profile as if it is a known fact, however a source is not quoted. Has Spacex said anything about the upcoming flight profile of SN9? To me, it makes no sense to simply repeat SN8 12.5 KM flight with the goal to “stick the landing”. The hardest part of doing things that have never been done before is: You don’t know what you don’t know. SN9 should enter a flight regime designed to teach them what they don’t know. “Sticking” the landing really isn’t that important, the flight test data is.

  11. When you say the rocket has 500,000 lbf of thrust. Do you mean at sea level or at altitude?

  12. I think there is a minor glitch with the new website. While on the upcoming launches page I can’t access this prelaunch preview but if I go to another launch, say turksat, it shows up as the latest prelaunch preview on the right hand side of the screen. If I click on that it brings me to this page, proving it does exist. Just trying to help

  13. I love how you precisely and correctly use metric system. Thrust or force in Newtons, mass in kilograms, weight (force) is not mixed with mass (amount of matter). Keep the good work!

  14. According to the definition of the words, this is both a launch and a flight, because a launch is when something is set in motion and a flight is when something flys, which this vehicle will do both. This is due to the way in which this rocket land and would not be an accurate description for all rocket tests.

    Was thinking about this and wanted to share

  15. Love your site and Youtube videos and Thanks for all the work you guys put in to it.

    I’m also catching the Space fever and made a site in Dutch about space, with a lot of links and Podcast ( a lot in english ) that i have found.
    Please visit my site and give your Comments please.

    Erwin from Holland

  16. I’ve done research and from what I can find NASA Russia China nobody has ever tried to use a gas turbine propeller or a turbine motor with propellers instance because they believe that it won’t work.
    you got to go by the basic law in theory which is you’ll (never know unless you try)
    by trying gives you more ideas on how you can improve it to make it work

    I’m 39 years old and I’m more than willing to try anything.. for instance if I was ever asked to take a 20-year trip in space I would do it and I wouldn’t even have to get paid for it I’d do it for the human race.. I am in going into space for 20 years even if they were to build some kind of habitat on Mars or the Moon and I would live out there for 20 years as experimental just so we can get all kinds of data and research and figure out more things I would do it build me a habitat on Mars give me the machinery and I’ll tell you 500 ft hole or even a thousand and ten thousand ft even though I’m not a scientist I still would do it hell even if they told me it was 40 years I would do it..

  17. Absolutely Great article here. I love the details of the events and how easy the lay out of the page is. This is exactly what I have been looking for! Keep up the amazing work!

  18. Why don’t they just land starships (that are not intended to land on other celestial objects) in horizontal position, without igniting its engines. I really like the idea of catching super heavy with hooks, arms or wires, thus removing functionality and weight from flight hardware and installing it on the ground (I had the same idea even before I heard about it). Here’s what would have to be bulit for starship landing infrastructure: some sort of reversed flame deflector, that creates strong enough upward wind, that the empty starship can comfortably slow down in it. And something to catch starship in that airflow. Essentially a big vertical wind tunnel (but open on the top).

    With dried out tanks, or tanks filled with just uncompressed, heated gases, the terminal velocity of the falling vehicle would decrease significantly (I assume) compared to SN8’s. The upward wind could be generated by rockets, turbines or even electric fans built around the flame deflector. The sound of landing might even become weaker. If you ever played with a ping-pong ball and hairdrier, you know that this isn’t even that hard of a physics problem. Tanking on orbit would go faster, if all the fuel could be transported (minus deorbit fuel). Raptors would have to burn less. Reigniting after flipping would not be neccessary. Flipping back would not be necessary.

    Probably more fuel (energy) would be used in total in order to accelerate that much air above the landing site, and sound pollution would last longer, but I feel like the benefits would outweigh the drawbacks.

    1. In case of 70m/s terminal velocity and 50m by 70m area, the power cost to create an airflow would be 1.44 GW (a raptor is 1.08GW). If velocity could be lowered to 50, and area to 40 by 20, the power could be 120 MW.

  19. As I have been watching these steps, one of the early ones is “Grass/Ground Venting.” Aside from venting in a particular location/elevation, what happens to SS during this step? Are they flushing lines? Pre-chilling tanks? It seems to take around an hour each time.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.