Starship Development History

It has almost been two years since the first hop from Starhopper took place with Raptor SN2, but we have seen SpaceX move at a rapid pace to develop their Starship prototypes. Here’s a recap of every single prototype, what they did, and how they ended their life, starting with Starhopper.


Starhopper was the first ever prototype made by SpaceX with 301 stainless steel. It was a proof of concept that steel could work for a craft of this size and it wouldn’t be a total mistake. Starhopper was tied down to the ground by a tether so it wouldn’t fly too high on it’s first attempt. On April 5th, 2019 Starhopper made its first hop. The prototype hit the end of the rope very quickly due to the engine, Raptor SN2, being at full thrust immediately. After this, SpaceX took the tether off and waited for another eng0ine to do a second hop. After installing Raptor SN6, on July 25th, 2019, Starhopper made it’s second hop, flying up to 20 meters and coming back down on the same spot. The flight was successful and Starhopper made its last and largest hop on August 27th, 2019, flying up to 150 meters, translating over and landing on the landing pad, where it still sits to this day. It was affectionately named the water tower that can fly.

Mark 1

Mark 1 was originally a display model of what the full starship would look like when completed. This was made in a race with the team down in Florida who was making Mark 2, a prototype which was scrapped and never saw the light of day. Mark 1 was placed on show next to the last Falcon 1 first stage to show the progress SpaceX has made from its first days. After the showing, the nosecone was taken off and Mark 1 was placed on a test stand for cryo testing and pressure testing. During the cryo test, Mark 1 popped its top and sent the forward dome flying hundreds of meters into the air, landing far away from the vehicle. The vehicle was declared destroyed.

Serial Number 1(Mark 3)

Serial Number 1, or SN1, was the first starship to have the new title of Serial Number. This prototype switched the manufacturing technique from bump-formed steel and flux core welding to stamped steel and TIP TIG welding, making the joints stronger and reducing the mass. Between the time this was announced and the actual rings were stacked, many prototype tanks were made and tested, like nose cone header and the main tanks. This prototype was designated to be “designed for orbit”, but it proved not to be. During the same cryo test that doomed Mark 1, a flaw in the thrust structure caused the tank to jump off the stand and crumple like an empty soda can when you step on it, and thus, SN1 was destroyed.


SN2 was a test tank of the main methane tank. It was testing if the cryogenic test was flawed or the prior tanks were flawed and this new design would work. In the end, SN2 was the first ever tank to pass a cryogenic test, although it never got a raptor engine due to it not being a full stack.


SN3 was another full stack in contingency with SN4. SN3 was supposed to be used for short hops eventually and was to be equipped with one Raptor engine. However, during a cryogenic test of the methane main tank, the oxygen main tank was filled with gaseous nitrogen instead of being used in the test. In the middle of the test, a valve at the top of the main oxygen tank failed, letting out all of the pressure from the tank. The rest of the stack could not support the weight of the methane tank and the liquid nitrogen inside, causing the bottom to crumple and fall over like an old Atlas rocket losing pressurization in its tanks. 


SN4 was originally designated to be a long flight model, going to be used for higher than 20 m hops. Because of the failure of SN3, SN4 was the first full stack to pass a cryogenic test. It used an augmented form of the cryogenic test that tested both tanks at the same time. SN4 also used a new quick disconnect system between the Ground Station Equipment and the test vehicle on the thrust puck between tests. SN4 was able to do two static fires out of both sets of tanks, main and header, and a new cryogenic pressure test was conducted. After a third static fire, the vibrations in the vehicle knocked the pipe that brings methane down to the thrust puck loose and caused methane to spill into the oxygen tank, lighting on fire and burning some of the insulation, causing significant damage to the lower part of the vehicle and not allowing SpaceX to control depressurization of the vehicle for 2 days. In the end, SN4 blew up after a fourth static fire after a failure with the GSE quick disconnect.


SN5 was the next vehicle after SN4. Elon Musk wanted to use this vehicle or its ancestor SN6 for an orbital test flight by the end of 2020. SN5 passed the cryo test and a static fire before its 150 meter hop. The 150 meter hop was the first of its kind, being the first full stack to leave the test stand under powered flight and the first full stack to land on the new landing pad after Starhopper’s inability to move led it to being stationary until this day. SN5 was recently disassembled and taken to the SpaceX scrapyard at the Boca Chica site.


SN6 was the next vehicle on the block and was meant to be a reassurance that SN5 was not a standalone event. SN6 went through all of the same procedures that SN5 did, up to and including the hop. In comparison, the hops looked almost identical, going up to the same height, going over the same distance, and being in the air for the same amount of time. SN6 was scrapped before SN5.


SN7 was a pioneering tank that looked at the use of 304L steel, a lighter and better steel than 301 steel. If SpaceX would be able to use this, it would cut down on cost and weight. SN7 was just a tank, but it was built in the same way as any other Starship prototype. SN7 was given a cryogenic test, then it was tested to destruction, holding 7.6 bar before popping its top.


SN7.1 was another 304L test tank that looked to reach a higher pressure than SN7. It was rolled out to the pad and was given the same tests as SN7. In the end, SN7.1 failed the pressure test at just under 8 bar, failing right where the dome of the tank and the sheet metal meet.


SN8 was the first full stack with a nose cone and flaps. Although it was planned to be fully built out of 304L steel, it is believed that some parts were still made out of 301 steel, and it cannot be proven if it was. This was the first proof-of-concept starship made, but the nose cone and flaps were installed at the pad after the first static fire. This test vehicle completed 3 cryogenic tests and 4 static fires between October and November 2020. After the 4th static fire, some of the debris on the pad caused SN8 to lose pneumatics. The original altitude of the high altitude flight test was 20 kilometers, but it was lowered to 15 km and then further lowered to 12.5 km. The first attempt for a launch ended in the first Raptor auto-abort of the program. The next attempt was the first launch of a full Starship test article. SN8 successfully launched off of Suborbital Pad A, made it up to its apogee, flipped into the belly flop maneuver, and fell right over the pad. However, the methane header tank unexpectedly lost pressure right before landing, which caused one of the two raptor engines to fail, sending all of the fuel from the header tanks through one engine, making it run “engine-rich”. “Engine-rich” exhaust happens when the pre burners start to burn through the metal in the engine, causing the metal to come through the engine bell. In the case of Raptor engines, copper started coming out the engine, turning the engine exhaust green, and causing the engine to lose thrust. This caused SN8 to slam into the landing pad and almost disintegrate into small pieces. 


SN9 was the second full stack. The methane header tank was switched from autogenous pressurization to nitrogen pressurization. SN9 was rolled out to the pad a few weeks after SN8’s failure, and was fully built out of 304L steel. However, SN9 was found to have broken its stand in the high bay and tilted, breaking the upper flaps. It took a while to put a new flap on, but they pulled one from the next SN and it was moved to the pad and placed on Suborbital Pad B. As SpaceX had a massive tangle with the FAA giving them clearance for the flight of SN9 after what happened to SN8, SN9 conducted 6 static fires in total, having 3 of them in one day, hours apart. Two raptor engines had to be switched out after the triple static fire. SN9 was scheduled to only hop to 10 km, and it successfully completed it with the same ease as SN8. However, as it was performing the flip maneuver, one of the engines failed to relight, causing the vehicle to lose control and hit the pad going way too fast to land. After this, Elon Musk commented that it was stupid not to use 3 engines for the flip.


SN7.2 was rolled out to a test stand as a test tank being used to test the viability of 3mm 304L steel instead of 4mm 304L steel. It was given one cryo test and one cryo pressure test. During the cryo pressure test, a small leak formed in a seal. SN7.2 has been on the test stand undergoing fixes to possibly go for another test.


SN10 was ready to go and rolled out to the pad before SN9’s launch. It was placed on the pad while SN9 was still undergoing tests, and even conducted the smaller test of seeing if it would survive the landing, or as it actually was, the explosion of SN9. SN10 had the most amount of heat shield tiles on any test vehicle up to its time. The tiles were being tested for how well they would do during the belly flop maneuver, but they were never able to see the results as the two previous Starships exploded. SN10 had only one cryo test and two static fires, the second one being needed after one of the Raptor engines was switched out after “looking suspect” by Elon Musk’s terms. After the switch, SN10 waited for a week and a half before launching, and launched up to 10 km. SN10 had some of the best footage of any launch and was the first Starship test article to land. However, when it landed, the thrust of the engine was low although being commanded high. This was the first issue of this type, being a major issue. This caused SN10 to crash into its landing legs at around 15 mph. SN10 jumped up after this, shearing the legs off and brought SN10 down to its skirt. The engines were buckled in and the steel rings buckled as well. The pipe bringing down methane broke, and methane leaked out. In the end, methane and oxygen combined near the hot engines, exploding out and destroying SN10. Elon Musk commented on the landing burn, saying they should use 2 engines for the landing burn from now on.


SN11 was rolled out onto the pad just five days after SN10’s landing and explosive aftermath. The difference in this rollout is that there were already 3 Raptor engines installed on the base of the Starship prototype. This will probably be used to test how Raptor engines handle transport on the base. All the Raptor engines were gimballed as far away from each other as possible. SN11 has yet to undergo testing and looks to have to wait a while before it gets any look at a flight. 

SN15 and onward

After SN12-14 were scrapped, SN15 was put up as the next Starship to go after SN11, as it has major changes from SN11. More starship models will continue to be made until they make a final design ready for human flight, which will probably take a couple of years. Starship and its testing site will continue to grow over a long time and it will be interesting to watch.

BN1 and onward

As the Orbital Launch Site looks to be getting ready, it seems that the first Superheavy booster is getting ready for its tests as a heavy duty transporter and the 304L steel rings are getting stacked for the proper move and test. Superheavy is the booster that will help Starship along, but it will be interesting to see how it develops and shows its path.

Editors Note: Since the writing of this article Super Heavy BN1 has been stacked and Elon Musk had announced that it is a pathfinder and will not be flown.

Starship is the rocket of the future and it is interesting to be able to see the development and testing of such a rocket.