Falcon 9 is the most-flown orbital launch vehicle in history. Its first-stage booster lands vertically on a drone ship or a concrete pad, gets refurbished, and flies again — in some cases more than 20 times. Dragon, the capsule Falcon 9 most often carries, is the only vehicle currently launching astronauts from U.S. soil. Together they define the commercial-space era.
How Falcon 9 Is Built
Falcon 9 is a two-stage, kerosene-oxygen rocket. Its first stage uses nine Merlin engines burning RP-1 (highly refined kerosene) and liquid oxygen, delivering roughly 7.6 million newtons of thrust at liftoff. The second stage uses a single vacuum-optimised Merlin. Both stages are made of aluminium-lithium alloy tanks with common bulkheads. The vehicle is deliberately simple in principle — the engines, the propellants, the structure are all conventional choices — but engineered to be produced and flown in volume.
The thing Falcon 9 did that nothing before it had done at scale was recover and reuse the first stage. After second-stage separation, the booster uses a combination of grid fins, cold-gas thrusters, and its own engines (throttled deeply and fired in a "boost-back" or "entry" burn, then again for landing) to return either to the launch site or to a remote-controlled drone ship downrange. The first successful orbital-class booster landing was in December 2015. As of 2025, SpaceX has completed hundreds of booster recoveries, and individual boosters have flown more than 20 missions each.
Falcon Heavy is a heavier variant: three Falcon 9 first stages strapped together, with the two outer boosters landing back at the launch site and the centre core either landing on a drone ship or being expended depending on the mission profile. Falcon Heavy launches rarely — most of its payload niche is now covered by Starship — but it remains the operational rocket of choice for heavier deep-space missions, including Europa Clipper in 2024.
Why Reuse Worked
Reusable rockets had been attempted since the Shuttle era without commercial success. Falcon 9 succeeded where others failed for a few specific reasons.
- Vertical landing under thrust. Vertical-landing propulsive recovery avoids the cross-range, thermal-protection, and refurbishment costs of Shuttle-style winged re-entry. The booster lands almost on its tail, with minimal heating and minimal airframe stress.
- Enough performance margin. Falcon 9's engines have enough thrust that the booster can afford to keep 5–10 percent of its propellant for the landing burns. Earlier reusable-rocket concepts were too close to marginal to pay that cost.
- A large captive market. SpaceX's own Starlink satellite constellation provides a steady pipeline of internal launches, which let the company fly Falcon 9 more often than any customer-only business model would have supported. Reuse makes sense only when you can amortise the first-stage refurbishment across many flights.
Dragon 2
Dragon 2 is SpaceX's current crew and cargo capsule, developed under NASA's Commercial Crew and Commercial Resupply programmes. It comes in two variants:
- Crew Dragon. Carries up to four astronauts to the International Space Station on NASA rotation missions, plus private-astronaut flights. Crew Dragon achieved its first crewed flight (Demo-2) in May 2020, the first crewed U.S. orbital launch since the Shuttle's retirement in 2011.
- Cargo Dragon. A pressurised cargo variant, externally nearly identical, carrying supplies and experiments to the ISS and returning downmass to Earth. It is the only ISS cargo vehicle that returns to Earth intact; all others (Cygnus, HTV, Progress) burn up on re-entry.
Dragon splashes down under parachutes in the Atlantic or Gulf of Mexico. Early Dragons splashed into the Pacific; the recovery zone was moved east to shorten crew-return times. Each capsule is reused across multiple flights, with the heat shield replaced between missions.
Dragon also performs in-flight abort capability via eight SuperDraco engines integrated into the capsule walls. In an anomaly during ascent, SuperDracos can fire and pull the capsule clear of the booster. The system was demonstrated in-flight with an intentional ascent abort in January 2020.
Why This Matters
Before Falcon 9, access to orbit was effectively priced like a government service: a few launches per year per national programme, at costs that were not declining. Falcon 9 flew over 130 times in 2024 alone, priced per launch in the $60–70M range for external customers and much less internally, with booster turnaround times measured in weeks rather than months. The commercial satellite industry, the current generation of lunar landers, Europa Clipper's flight path, and the entire Starlink constellation all exist in roughly their current shape because Falcon 9 made them affordable.
Falcon 9 is not the future — SpaceX itself intends Starship to eventually replace it. But it is the present. Every satellite operator, every science mission team, and every commercial space station programme currently in development is planning around Falcon 9's capabilities as a baseline. See the Starship article for the vehicle intended to succeed it, and the Orbital Habitats article for where Dragon fits in the ISS and post-ISS economy.