Opinion: Is the Starship flight test a reality check for SpaceX, NASA and Artemis?
By all honest measures, SpaceX’s first integrated flight test of Starship was a success, but it unearthed some serious problems.
Yes, that pun was intended.
When the Starship’s Super Heavy booster ignited its 33 Raptor engines and spooled up to some 16 million pounds of thrust, the concrete and ground below couldn’t handle the pressure. As the rocket began lifting off the pad at 9:33 a.m. EDT (13:33 UTC) April 20 at SpaceX’s Starbase facility in South Texas, it carved a massive crater below the launch mount. More on that later.
This was the first integrated flight test of SpaceX’s 394-foot (120-meter) tall two-stage Starship launch vehicle — the largest and most-powerful rocket ever to soar skyward. The flight was aspirationally expected to take the upper stage “Ship” around the planet to reenter over Hawaii.
This, of course, did not happen as the rocket began experiencing problems almost immediately after it left the pad.
At least three methane-oxygen-consuming Raptor engines were out within seconds of liftoff. It’s unclear whether they were lit at all, or if flying debris from beneath the rocket knocked them out. But as the rocket soared skyward, there were several more engines (maybe up to eight) that flamed out — some explosively — and it is possible that it lost hydraulic power to gimbal the engines at some point, although that has not been confirmed.
Ultimately the rocket began to tumble between two and three minutes into the flight and stage separation did not occur at all when it was supposed to around three minutes after liftoff. It also appeared, based on the SpaceX webcast, that the booster’s engines, or at least some of them, did not shut down when expected.
Finally, about four minutes after launch, Starship exploded over the Gulf of Mexico after the rocket’s flight termination system was activated.
It’s important to note that this was a test flight in a rapid-iteration test flight program. There are more Super Heavy boosters and Ships waiting down the road. Explosions and failures are anticipated. Expectations were justifiably set very low by the company.
The rocket managed to clear the launch pad and move eastward well out over the water, which was the most important box for this test flight to check. Based on the company’s webcast, the rocket made it to a maximum altitude of about 24 miles (39 kilometers) with a velocity of 1,330 miles (2,150 kilometers) per hour.
All things considered, this was a big win for SpaceX. As tests go, this was a success and valuable data was gathered by this launch.
But more than one truth can coexist.
This flight showed the Starship system has a long way to go in terms of development time before it’s ready to begin sending anything usable into orbit, and even longer before it’s ready to serve as a lunar lander for NASA’s Artemis program to return humans to the surface of the Moon.
Should this be a reality check for NASA’s currently publicized timelines for sending humans to the lunar surface during the Artemis 3 mission?
NASA has contracted SpaceX to develop a lunar variant of the Starship vehicle, which would act as the lunar landers for both Artemis 3 and Artemis 4. According to the U.S. space agency, the contract for both of these missions is worth $2.89 billion and $1.15 billion, respectively.
Under the current timeline, NASA hopes to launch the Artemis 3 Space Launch System rocket with four astronauts no earlier than late 2025. Even before taking into account Starship development, this is very optimistic.
Moreover, it is possible the spacesuits for moonwalking, being developed by Axiom Space, won’t be ready in that timeframe, either.
And if the development schedule wasn’t already tight, the impact of Starship’s first integrated flight test on the launch infrastructure will all but guarantee the system won’t be ready to support Artemis 3 by 2026.
Starship one, ‘stage 0’
After the launch, it has become abundantly clear that there will have to be some changes to what SpaceX calls “stage 0” — the launch pad and its infrastructure — changes that could take some time.
Part of the reason it was so important for Starship to clear the tower and get out over water was to avoid any significant launch site damage should the rocket explode or fall back soon after liftoff.
But it turns out that the way the pad was designed — without a flame diverter or significant water deluge system — ended up allowing the rocket’s engines to dig a crater below the launch mount, causing a multitude of problems.
Starship’s liftoff energy had to go somewhere, and because there was nothing directing it in a specific direction, the energy and the excavated debris went in all directions.
Liftoff from Starbase pic.twitter.com/rgpc2XO7Z9
— SpaceX (@SpaceX) April 20, 2023
In a wide drone shot provided by SpaceX, as Starship was launching, massive chunks of debris — possibly concrete, dirt, shrapnel or some combination — were seen flying up past the vehicle and pummeling the surrounding area.
There were even very visible large splashes in the Gulf of Mexico at least 3,000 feet (900 meters) away from the launch pad moments after the vehicle began rising into the sky.
In aircraft flyovers of the launch pad later in the day, imagery showed a hole at least 40 feet (12 meters) deep was carved by the 30+ Raptor engines, which exposed parts of the launch mount’s foundation.
Company CEO Elon Musk said all that was left of one of the concrete lateral support beams was rebar. It’s unclear if that can be fixed without completely rebuilding the launch mount.
Moreover, there were dents in the nearby tank farm from flying chunks of debris. Perhaps those will buff right out?
Does that launch time include repairs to other parts of Starbase that withstood damage? pic.twitter.com/E1FNwwkZAL
— Johnny Beatty (@Mr__Beatty) April 21, 2023
Years ago, Musk tweeted that he was aspiring to have no flame diverter for Starship, “but this could turn out to be a mistake.”
In a tweet after images of the pad damage began circulating, Musk said SpaceX, three months ago, started building a “massive water-cooled, steel plate” that would go under the launch mount.
“Wasn’t ready in time & we wrongly thought, based on static fire data, that Fondag would make it through 1 launch,” Musk said. Fondag is a type of high strength concrete.
That static fire he is likely referring to occurred in February and saw 31 of 33 Super Heavy booster engines fire at 50% of rated thrust for just five or six seconds. The actual launch attempt had at least 30 engines firing at full thrust (16 million pounds) for nearly 15 seconds at or near the pad.
“Still early in analysis, but the force of the engines when they throttled up may have shattered the concrete, rather than simply eroding it,” Musk tweeted. “The engines were only at half thrust for the static fire test.”
Stage 0? How about ground zero??@SpaceX a suggestion: create a MOAT around stage 0 (Mega Oceanic Acoustic Tamper): a giant 15m deep tub (outside the main structure here) that's filled with natural water (and/or fresh water) with a metal surround–the tub 2x the size of the posts… pic.twitter.com/aTCi8gwiaz
— DrKnowItAll (@DrKnowItAll16) April 21, 2023
Musk says the pad can be ready to launch again in a month or two, but he is known to make wildly optimistic timeline predictions. Don’t be surprised if it takes at least six months or more, minimum, especially if a massive overhaul is required to ensure the integrity of the launch mount.
While SpaceX hasn’t said what caused the initial three Raptor engines to fail, it’s entirely possible that debris and acoustics from the rapid unscheduled digging event damaged parts of the Super Heavy booster. Would Starship have made it through stage separation and beyond had there been a proper flame diverter and water deluge system?
Criticism aside, this is a test program. The goal is to gather data. Sometimes the data tells you that you made poor design choices. This is usually a good thing, as long as you learn from it.
Starship liftoff in slow motion pic.twitter.com/KqHjqwP88Z
— Elon Musk (@elonmusk) April 22, 2023
Is it time to reset schedules and expectations?
Once SpaceX gets the pad and surrounding infrastructure fixed or redesigned, however long that takes, Starship still needs to get into orbit — something increasingly questionable to occur in 2023.
Then the heat shield on the “Ship” needs to be evaluated via an actual reentry profile, which can’t happen until Super Heavy and its Raptor engines can reliably reach stage separation — something this first flight has revealed also needs more work.
It doesn’t get easier from there. Both Super Heavy and the Ship will need to be reliably recovered — this is a reusable rocket system, after all. And the former is expected to be caught using catch arms on the launch tower — something so crazy it might just work.
After that, SpaceX still needs to develop a tanker variant of Starship, prove out in-space cryogenic propellant transfer (something that’s never been done before) and build an in-space depot that has minimal propellant boil off (also a spaceflight first).
While all of that is ongoing, a lunar Starship prototype will also have to be built, flown and flight tested on the Moon well in advance of a crewed Artemis landing before doing it all over again for the crewed landing.
That is a lot to do in 2-3 years — perhaps an impossible amount. This isn’t a criticism of SpaceX. Developing a system like Starship with all of its promises is hard. Things will need to be tested, broken and rebuilt.
Sometimes things just take time — even before “oopses” like accidentally using your rocket to start digging a flame trench.
As such, this may be a reality check for NASA and SpaceX, suggesting that humans won’t be landing on the Moon by 2026 and possibly won’t do so until 2028.
The Artemis 2 free-return flight around the Moon is expected by late 2024 or early 2025. The crew for that mission was just announced earlier this month and will only require SLS and Orion.
If NASA continues to manifest Artemis 3 as a surface mission and it doesn’t fly until 2028, that would be a gap of three or four years between it and Artemis 2.
NASA plans to award a contract to a company other than SpaceX for a second lunar lander to be developed alongside Starship to provide the agency with dissimilar redundancy in case one lander has problems.
But as far “behind” this very aspirational timeline as SpaceX is, its lander design is still likely to be the only one ready to land astronauts by 2028.
Perhaps Artemis 3 should be re-manifested as a mission to the lunar Gateway, which should be circling the Moon by 2026. Then Artemis 4, which won’t be ready to launch with an upgraded SLS Block 1B until 2027 or 2028 anyway, can serve as humanity’s return to the Moon.
The first integrated test flight of Starship was a great proof of concept for the entire system. It will eventually work and when it does, it will open up deep space — the Moon, Mars and beyond — in ways not possible today with anything else even on the drawing board. It will just take more time than advertised.
The opinions expressed above are solely those of the author and do not necessarily reflect those of Spaceflight Insider
Video courtesy of SpaceX
Derek Richardson has a degree in mass media, with an emphasis in contemporary journalism, from Washburn University in Topeka, Kansas. While at Washburn, he was the managing editor of the student run newspaper, the Washburn Review. He also has a website about human spaceflight called Orbital Velocity. You can find him on twitter @TheSpaceWriter.
Still a successful flight! Just getting it off the ground with the most powerful and tallest rocket is an accomplishment. Knowing Musk, he will get this problem right! Made in the USA!
As someone who worked on F9, thank you for balanced and well-researched reporting on the Starship test. I’ve been disappointed by so many articles that are either all-praise or all-doom, with tons of factual errors. Space is hard, is going to remain hard, and requires hard work to come. This is an important reminder, and also comes with useful data.
NASA was supposed to have at least two other private commercial options besides the Space X Starship for reaching the lunar surface. But instead it reverted back to giving just one company the contract.
Never put all of your eggs in just one basket.
I fully concur with a previous commenter regarding the need for informed and balanced reporting such as this. Even some of the more knowledgeable YouTube channel hosts who make a living covering SpaceX’s progress have only been concerned with speculative failure analysis, not the realistic impacts to NASA timelines going forward.
The thing folk are overlooking here is that on the 20th of April SpaceX launched the largest rocket ever built without a properly functioning flight termination system.
The FTS punched holes in fuel tanks but detonation of the fuel only happened when the ship collapsed 50 seconds later. This was fine at an altitude of 39 kilometers but had the loss of control happened only a few kilometers off the ground, 50 seconds would have seemed like an eternity.
However, a few kilometers off the ground the tanks would have had enough fuel to maintain the ships structural integrity for way longer than 50 seconds. If the activation of the FTS had happened a few kilometers off the ground, the only thing it would have achieved is turning the rocket into a slightly more effective fuel air bomb upon impact with the ground.
Artemis 3 is still scheduled for 2025 Not 2026 the next starship should fly in under 6 months and that will certainly make it to orbit.
This test flight showed the vehicle design is very robust, the stage zero performed perfectly aside from the erosion and crater damage it’s clear it would have succeeded.
In orbit propellant transfer is easy compared to launch or landing.
Once in orbit there are no time constraints if a transfer takes days or has more boiloff than expected no problem send an extra tanker up – it’s just repeating the same thing again.
What will happen when the Starship lands on the Moon or Mars, kkeping in mind what happened to the launch pad , surrounding areas and the vehicle on this first integrated flight test? How about if it ever lands on the surface of the Moon or Mars and tries to take off? It seems that some crucial things weren’t thought out adequately. Good luck on this venture. I was pretty optimistic until this thing happened.