After two Starship prototypes achieved launch debuts in the height only to lose landing errors at the last second for unique reasons, SpaceX is planning for a third launch already this week.
Crucially, although Elon Musk’s high-level realism (or pessimism) has often seemed to underestimate the actual odds of success, the SpaceX boss is significantly more confident in Starship’s third launch attempt than he was on the first flight two months ago. Back when Starship serial number 8 (SN8) was preparing to try out the program’s first high launch, Musk fixed the probability of a successful launch, free fall and landing at ~ 33%.
As it turned out, he was not wrong, but Starship SN8 did in the end far closer to a total success than almost anyone ̵
In other words, both float tests served their nominal purpose and revealed two error modes that would eventually have lifted the head in some way. With the SN8, Starship failed to maintain enough pressure in its secondary landing gas tank to supply two Raptor engines with enough fuel for a landing combustion. Starship SN9 failed a few seconds before SN8 when one of the two Raptor engines needed for a flip and landing combustion never ignited, causing the rocket to crash into the ground at an angle relative to SN8’s impact.
As previously discussed at Teslarati, Elon Musk eventually revealed his opinion that SN9’s engine output failure could potentially be avoided, and that SpaceX would change the way future Starships try to land in an attempt to add more redundancy.
While SpaceX has obviously not spun around and solved a complex Starship propulsion problem in a matter of days, Musk finally revealed his opinion that he, his engineers, or a combination of both “were too stupid” to exploit an obvious way to reduce the risk of engine failure below [Starship SN9’s] flip and landing. The ‘obvious’ refinement: re-ignite all three Starships available landing engines, not just two. ”
By igniting not just two – but all three – Raptor engines during Starship’s flip burn, SpaceX could essentially perform a static fire in the air, giving the rocket’s flight computer a few seconds to analyze the performance and fallout of the two healthiest engines the final landing combustion. With that change implemented, Starship would theoretically have enough redundancy to land if only two of its three sea level Raptors performed nominally.
Starship SN10 is currently installed on one of two ‘suborbital racks’ on SpaceX’s South Texas launch pad, and will be the first high-altitude prototype to attempt the three-engine flip burn and on-the-fly downgrade. Musk says that confidence that SN10 will successfully land is now 60%, an almost double improvement compared to SN8. The Starship SN10 could potentially fly as early as this week, although the prototype still needs to complete a nominal static fire test with three engines, and the launch has not yet received FAA approval.
Further down the road, Musk says that SpaceX is working hard to improve the Raptors’ deep gas performance, possibly allowing future Starships to burn two – or even three – engines all the way to touchdown for even more redundancy. However, deep throttling of large, complex rocket engines is extremely difficult, so upgrading is likely to be no less than several months away. Meanwhile, Starship SN11 is effectively complete, and Starship SN15 through SN18 is brought together to support a relentless float test campaign while SpaceX works towards orbital flights.