HAWTHORNE, Calif., July 27 — Twenty-five years after the first generation of rocket engines that relied on simple gas-generator cycles, a new breed of powerplant is rewriting the rules of launch economics. The Raptor engine, built by SpaceX to propel the Starship system, is the first full-flow staged-combustion engine ever to reach flight, and its numbers are as audacious as the vehicle it pushes off the pad. The Raptor doesn’t just burn propellant — it routes both liquid methane and liquid oxygen through separate preburners before the main combustion chamber.
Think of it like a two-way highway where both fuels get their own dedicated acceleration lane, squeezing more efficiency out of every drop. The result is a sea‑level thrust of 510,000 pounds‑force per engine.
Stack 33 of those under the Super Heavy booster, and you get a staggering 16.7 million pounds‑force at liftoff — enough to loft a 395‑foot‑tall vehicle that weighs as much as a fully loaded office tower. The upper stage, Starship itself, stands 165 feet tall and uses six Raptors: three tuned for vacuum operation and three for sea level. Their combined vacuum thrust reaches about 5.1 million pounds‑force.
In vacuum the engine achieves a specific impulse of 350 seconds; at sea level the number drops to 330 seconds. For context, that vacuum figure is better than any kerosene‑burning engine flying today, and it’s what makes the Starship system capable of lifting 100 to 150 metric tons to low Earth orbit in reusable mode — or up to 250 tons if SpaceX decides to expend the hardware.
Learning to land
Getting an engine this powerful is one challenge; getting it to survive reentry and land again is another. Through the first 13 integrated Starship flights — the last of which occurred in early 2025 — SpaceX has logged a mixed but improving record. Four of those flights are considered full successes: the Super Heavy booster returned to the launch tower for a catch, and the Starship upper stage executed a controlled landing or splashdown.
Three more flights achieved a booster catch but lost the Starship stage, while the first six flights ended in failure. The booster’s catch success rate since the first attempt on Flight 5 stands at four out of nine attempts.
The upper stage has fared better in one respect: it survived reentry intact on eight of the 13 flights (Flights 4, 5, 8, 9, 10, 11, 12, and 13). Those numbers point to an engine and a vehicle that are gradually proving their mettle under extreme conditions — the kind of iterative improvement that SpaceX has turned into a hallmark of its engineering culture.
The dollars and sense
What makes the Raptor more than just a technical marvel is its price tag. SpaceX has driven the manufacturing cost of each engine below $1 million, a fraction of what comparable high‑performance engines historically cost. That frugality flows directly into the company’s target of $10 million per flight — a figure that, if achieved, would slash the cost of putting a satellite in orbit by an order of magnitude or more.
The practical upshot: Starlink, SpaceX’s satellite‑internet constellation, generated $4.2 billion in annual revenue as of 2024, and it is the primary customer waiting for Starship’s mass‑deployment capability. A cheap, high‑cadence launcher means SpaceX can replenish its orbital fleet faster and cheaper than competitors can even launch a single batch.
The company has stated it aims eventually to fly each Starship up to three times per day. The Raptor engine — along with the full Starship stack — keeps pushing toward that daily‑flight cadence. With each booster catch and each upper‑stage reentry, the gap between audacious design and routine operations narrows a little more.
