During a of engine tests of 's Dran this past Saturday, the vehicle eperienced what the has characterized as an "anomaly." Based upon an unauthorized leaked video of the accident, the was counting down towd a firing of the Dran's SuperDraco thrters when the vehicle eploded.

has not validated the , but is consistent wh verbal accounts of the flure that have been shed wh s.

After the accident, lge tic clouds of orange billowed above Landing Zone 1, w conducted Saturday's engine tests. According to one source, the orange plumes were the result of between one and two tons of nrogen tetroide—the oidizer ed by Dran's SuperDraco engines—burning at the location. After a tic weekend, what follows is a summy of what we k, what we k, and w es from .

What was destroyed?

The capsule in question is the same one that successfully flew a demonstration mission to the International Station in Mch. The was being preped for a launch abort test this . During this test, the Dran would have launched from Florida on a 9 booster and then d s powerful SuperDraco engines to show that the Dran could pull self ly away from the , in case of a problem wh the booster before or during f.

that has lost this capsule, mt find a substute for this launch abort test. is not cle whether fabricate a boilerplate vehicle wh a SuperDraco sy of eight thrters, or repur one of the Drans has built for crewed fs to the station. Eher way, this is a signicant materiel loss for the .

How did hen?

We k. According to the leaked , the anomaly occur whin the final 10 seconds of the countdown, and is not entirely cle whether the SuperDraco engines had begun to . One source indicated that the has a lot of about the flure—this was a ground-based test, so the vehicle was heavily instrumented—so theoreticy finding the root cae of the accident should be strghtforwd than a problem had occur during a real f. The best-case scenio, in terms of caing s for , would be that someone mishandled the ground sys equipment. The worst-case scenio is that is some undiscove but funental problem in SuperDraco thrters.

During past , has been frly forthcoming about the cae of the flures, and we hope for simil transpency wh this accident. I would gue that, since this vehicle eventuy cry s and is funded lgely by , transpency is essential to ensuring confidence in the vehicle and 's processes.

Was anyone ?

Thankfully no. The last we saw this tic of a ground-based flure was during the Apollo 1 fire in 1967, which cost three lives. Fortunately, no one was hmed during Saturday's accident, which speaks well of 's prices during such dynamic tests. Had s been inju or killed, would have undoubtedly complicated the already comple road a for .

What does this mean for fs?

provided multibillion-doll contrs to and in 2014, wh the intent of bringing their Dran and Stliner vehicles into service for getting . to the station. Before this accident, and had been tgeting ely October for the crewed Dran mission to the station. , that almost certnly be ed by at least several months, into 2020. Before Saturday, 's Stliner was behind Dran in terms of development, and is also unlikely to fly s before ely 2020.

recently signed a deal wh to purchase two addional Soyuz ts, for one crew member each, which ensure a crew presence on the station through September 2020. The ncy well be forced to return to the ns yet agn to procure ts through the end of 2020.

What does do ?

Undoubtedly, the had a by Easter weekend. The step is to determine what hened and then wh to fully understand the problem; they would then devise a fi to ensure the problem never hens agn. Interny, the 's engineers already k what occur.

I would also be hugging were I in manment—leaning on the ncy for s epertise in f sys as well as seeking cover from polical . After a 9 launch flure in 2015, in which the CRS-7 supply mission to the International Station was lost, the ncy stood by s commercial c ptner. f iam Gerstenmer offe support for the , beat back congressional doubters, and ed get back to flying quickly.

In recent yes, some have ed the ncy as "hing back" during the development of the vehicle wh unnecessy paper and requirements. This be ptly true, but is the ctomer, and clely e hazds yet to be found in the Dran (and probably Stliner too). The is that needs to succeed, and so the and the ncy e presently in a posion w 's best for everyone they together by , identy and fi the issue, and move on.

is a precedent for this. After the Apollo 1 fire revealed multiple problems wh the version of the , ed closely wh the capsule's contr, North ( a pt of ) to accelerate of a much r updated capsule . The occur in Januy 1967, and the updated "Block II" capsule made s f less than 21 months later. The would on to fly a historic ion of lun missions.


would be easy to wre off as a reckless . But the realy is that this is a rapidly in a lot of dferent directions—building the 's lgest operational ( Heavy), perfecting st ree, launching s than any , trying to recover payload frings, and building an unprecedented, net-generation vehicle ced Starship.

This accident should offer a clying for and Mk that rey mt get right—and that putting s on a 9 , in a Dran , rses the stakes. This is not easy. is very hd.

should be ltle doubt the can come back from this. has shown a propensy for responding to flures wh speed and an abily to fi problems. After the CRS-7 flure in 2015, they were flying agn si months later. Remkably, the return to f mission also was the first successful 9 landing.

After the Amos-6 launch pad flure in 2016, the was flying agn less than five months later and has had s most ful run since then. The can get beyond this accident, but that s e involved require foc, transpency, and closely ing wh to move on.

This story originally appeared on Ars Technica.

Related posts