SpaceX’s first attempt at landing on land was also its critical return-to-flight mission after a failure just months before. The hypersonic grid fins—unique among orbital class rockets today—helps stabilizes the rocket during re-entry. Credit: SpaceX
One year ago, while climbing toward maximum dynamic pressure, a SpaceX Falcon 9 rocket erupted into a cloud of debris. Several months later, it was determined that a strut supporting a helium tank had ruptured mid-flight, turning the tank into a pinball.
The unmanned Dragon cargo capsule carried by Falcon survived the rocket’s breakup, but no system existed to deploy the parachutes before Dragon smacked into the Atlantic, destroying the cargo.
On Tuesday, April 14, 2015, SpaceX Falcon 9’s first stage attempted a precision landing on our autonomous spaceport drone ship named “Just Read the Instructions”. The stage made it to the drone ship and landed, but excess lateral velocity caused it to tip over. Credit: SpaceX
Recent history suggests that catastrophic rocket failures like this, which occur during the Stage 1 portion of flight (or during the transition between Stages 1 and 2), are difficult to recover from. Since 2000, at least 14 orbital class rockets have experienced such failures, and the average time before returning to flight was 447 days. After its Falcon 9 failure, SpaceX resumed flights just 177 days later—less than half the average delay.
In addition to its fast recovery from failure, SpaceX has recently deployed and tested numerous vehicle upgrades and hardware modifications with incredible turnaround time. When Falcon 9 began flying in 2010, it had a shorter airframe, less efficient engine configuration, and lacked landing legs and hypersonic grid fins. These changes have given Falcon 9 a dramatically enhanced payload capacity and enabled (sometimes) successful landings at sea. SpaceX’s apparent willingness to fail, especially on its experimental landing efforts, looms large in enabling these hardware upgrades to come online.
Sped up video of the Falcon 9 first-stage landing during the THAICOM 8 mission on May 27, 2016. Credit: SpaceX
Contrast SpaceX’s efforts in testing its human-rated Dragon V2 capsule with that of Lockheed Martin’s Orion Capsule. Lockheed tested Orion’s launch-abort systems in a pad abort test on May 6, 2010, while the first flight of Orion took place more than four years later, on December 4, 2014. SpaceX conducted a pad abort test of Dragon V2 on May 6, 2015, and has scheduled an uncrewed orbital flight for May 2017. Even if that test flight is delayed by a period of one or two years, it will occur with faster turnaround time than the Orion tests.
The abort methodology, too, warrants consideration. While Orion relies on a “tower” system previously used by Apollo capsules unboard the Saturn V, Dragon V2 has built its escape systems into the capsule body itself. These rocket motors, if unused during a successful launch, can be deployed to land the craft propulsively—on Earth or perhaps on another space body. This is a capability that Orion lacks. Thus, SpaceX is arguably testing a more complex, innovative abort system, and attempting to do so in half the time.
An artist’s concept image of the Orion capsule. Credit: Lockheed Martin
Another contrast of note is hardware modification in the event of a design failure. In October 2014, Orbital ATK experienced the failure of its Antares rocket, powered by NK-33 engines, just a few seconds off the pad. A review recommended swapping these engines in favor of the more reliable RD-181 (a swap that had already been planned, but was accelerated after the accident). Orbital is scheduled to return Antares to flight in August 2016 with the new engine configuration—nearly two years later.
By comparison, SpaceX first attempted a controlled descent onto the open ocean using its partially spent first stage and experienced an uncontrolled spin exacerbated by centrifuging of propellant. Less than a year and a half later, SpaceX instituted a hardware fix by adding hypersonic grid fins to the first stage rocket body.
The rapid pace of testing may have downside risks. Of the 14 orbital class Stage 1 failures, three were on SpaceX flights (the aforementioned Falcon 9, and two Stage 1 Falcon 1 failures). No other organization experienced as many Stage 1 failures in the same period.
The rapid pace of testing, recovery from failure, and willingness to fail in the first place helps make SpaceX arguably the most agile of rocket and spacecraft companies. The parachutes that failed to rescue Dragon after the rocket breakup? SpaceX added failsafe software to engage them in the event of a similar failure in the future.