Elon Musk’s plans to go to Mars next year are toast

SPACEX’S NINTH test flight of its Starship launch system could be counted as an advance over the previous two, which lit up the Caribbean skies a couple of thousand kilometres down-range of their Texas launch site like spectacular fireworks. A second stage successfully turned off its engines after reaching space intact, but by the time it re-entered the atmosphere on the other side of the world it was clear that this was a third failure in a row.

The main goals of the latest test flight, launched in the evening of May 27th, were to show that the system’s second stage—the pointy bit which is the Starship proper—could relight one of its engines once in space and to gather reams of data from its controlled re-entry so as to improve the heat shielding on future craft. Unfortunately, when the Starship turned off its engines after its initial ascent it found itself rolling, and its Earthbound controllers could not get it to stop. The engine relight was cancelled, all remaining propellant was vented from the tanks and the Starship plunged to fiery oblivion over the empty part of the Indian Ocean where it had been meant to make a controlled splashdown.

This Starship, like those which flew in the previous two flights, was of a new and supposedly improved design compared with the original version used in the first six flights. Its “Block 2” design’s purported enhancements—including higher thrust and fins better suited to re-entry—have not yet had a chance to manifest themselves. The tendency towards fire in the engine bay which doomed the previous two flights seems to have been sorted out, but a loss of fluids from one of the redesigned propellant tanks proved just as fatal. It is possible that this new problem, like the old one, stemmed from excessive vibration.

The issues may be sorted out through modifications to the remaining Block 2 ships; they may inform better design choices in the Block 3 ships. They may also indicate that the goal of a regularly reusable ship so light in its structure that it can get 150 tonnes into orbit is overly ambitious, and that the payload capacity will need to be reduced.

Regardless of what happens next, the three failures mean that the company is no nearer getting a Starship into orbit and successfully back to Earth than it was six months ago, when the last Block 1 Starship managed a controlled splashdown. It has no new data on how best to cope with re-entry; it has no new experience turning engines on and off in space; and it is no longer building the sort of Starships which previously provided it with such things.

The lost six months confirm that Elon Musk, SpaceX’s boss, will not be able to send uncrewed Starships to Mars next year, as he had claimed he intended to. The problems also cast serious doubt on the company’s ability to land an uncrewed Starship on the Moon next year as a precursor to delivering a crew there the year after as part of NASA’s “Artemis” programme. Starship is not the only part of the programme behind schedule; the 2027 date is hard to credit.

The losses have also pushed back the day when Starships might take over the job of launching the Starlink satellites which provide SpaceX with the bulk of its revenue. SpaceX’s Falcon 9 launchers are on track to launch more Starlinks this year than ever before, but the service is still unable to provide bandwidth to all those who want it in some areas. In principle Starship will be able to increase the rate at which capacity is added. And the fact that the most recent test flight successfully reused a previously flown “super heavy” booster to speed the ill-fated Starship on its way was a hopeful sign that such launches can one day become routine. But that day seems to be getting no closer. ■