---
title: "SpaceX Catches Starship From Orbit, but Full Reuse Still Waits on the Heat Shield"
description: "Ship 42 came back from 14 orbits intact, but no Starship has flown twice and the cheap-launch math depends on doing that often."
author: "rews special report"
published: 2026-11-15T00:40:59Z
modified: 2026-11-15T13:06:15Z
url: https://rews.cc/a/spacex-puts-starship-in-orbit-again-but-its-booster-ends-up--9f5b8f
language: en
type: special report
tags: ["spacex", "space", "starship", "reusability", "starlink", "tech"]
publisher: "Rews (https://rews.cc)"
---

# SpaceX Catches Starship From Orbit, but Full Reuse Still Waits on the Heat Shield

*Ship 42 came back from 14 orbits intact, but no Starship has flown twice and the cheap-launch math depends on doing that often.*

Special report · By rews special report · November 15, 2026 · https://rews.cc/a/spacex-puts-starship-in-orbit-again-but-its-booster-ends-up--9f5b8f

## In brief

- SpaceX caught Ship 42 with its Starbase tower on Nov. 14 after 14 orbits, the first Starship upper stage recovered from orbit
- Super Heavy Booster 22 cut its boostback burn short and was destroyed over the Gulf of Mexico, and the cause is under investigation
- No Starship upper stage has flown twice, and the two reflown boosters waited 131 and 221 days between flights
- Former NASA specialists say the tile heat shield cannot support rapid reuse, while SpaceX says Ship 42’s shield worked well
- FAA meetings on a plan for up to 456 Starbase launches a year run Nov. 17 to 19, with comments due Nov. 30

SpaceX’s launch tower at Starbase, Texas, caught a Starship upper stage in its mechanical arms on Saturday afternoon, Nov. 14, about 22 hours after the ship left the pad. It was the first time the company had recovered the vehicle intact after a trip to orbit.

The ship, Ship 42, circled Earth 14 times and released 26 Starlink V3 satellites. Then it dropped through the clouds and settled between the tower’s “chopstick” arms above Launch Pad 2, [The Associated Press and Reuters reported](https://aljazeera.com/news/2026/11/15/spacexs-starship-captured-by-launchpad-arms-in-dramatic-first). Its booster never came home. On Friday the Super Heavy had been diverted over the Gulf of Mexico and destroyed, so the catch of both stages that SpaceX promised on its October earnings call didn’t happen.

The catch settles one question: a Starship can come back from orbit in good enough shape for the tower to hold it. It leaves open the questions that decide whether the rocket becomes the cheap, quickly reflown machine that Elon Musk, SpaceX’s founder, has described for years. No Starship upper stage has flown twice. The only two Super Heavy boosters ever reflown waited 131 and 221 days between flights, and no booster of the current Version 3 design has been caught. Three former NASA specialists say the tile heat shield can’t support rapid reuse. The orbital refueling test that NASA’s moon landings depend on has slipped from March 2025 to early next year. And the low per-kilogram costs attached to Starship are SpaceX’s internal targets, which rest on reuse the company hasn’t yet done.

## Two days at Starbase

Flight 15 lifted off at 4:47 p.m. local time on Friday, Nov. 13. It was the 15th full-scale Starship launch since 2023 and only the second to reach orbit. The first, on Sept. 28, ended after fewer than two laps when an engine failure forced SpaceX to bring the ship down early, [as rews reported](https://rews.cc/a/starship-finally-goes-orbital-drops-26-starlinks-and-clocks--13cd95). This time the ship flew at about 165 miles. Roughly 40 minutes after launch it pushed its satellites, each with solar arrays that stretch 60 meters end to end, out of a narrow door at its base, [Eric Berger, the senior space editor at Ars Technica, wrote](https://arstechnica.com/space/2026/11/starship-flight-15-powers-to-orbit-setting-up-a-possible-catch-on-saturday/).

[![SpaceX makes history with Starship Flight 15, catching rocket's upper stage from orbit](https://i.ytimg.com/vi/hVip4Von6qA/hqdefault.jpg)](https://www.youtube.com/watch?v=hVip4Von6qA)

A WESH 2 News report on Starship Flight 15 and the first tower catch of the rocket’s upper stage. Video: WESH 2 News · YouTube

Saturday’s return took less than a minute from the moment the ship came out of the cloud deck. Mr. Berger watched both days from Starbase. He wrote that the ship fell flat until the last moment, lit three Raptor engines, swung upright and eased toward arms that had only two nubs to grip, each sticking out about a foot. [Payload put the catch](https://payloadspace.com/spacex-successfully-catches-starship-upper-stage/) at 4:20 p.m. Eastern time, after one last swing over the Gulf.

> SpaceX has dazzled us before, by landing rockets on barges and returning humans to space with Crew Dragon. But this flight took things to another level.
>
> — Eric Berger, senior space editor at Ars Technica, [arstechnica.com](https://arstechnica.com/space/2026/11/with-starship-catch-spacex-still-shows-it-can-dazzle-the-world/)

Dan Huot, a SpaceX communications manager calling the webcast, told viewers, “We caught a Starship,” and then, “And now everything changes.” That evening Jared Isaacman, the NASA administrator, posted his congratulations. His agency is paying SpaceX billions of dollars to land Artemis astronauts on the moon with Starship.

> Congrats to @elonmusk and the @SpaceX team! It was almost magic..Making more believers by the day....one step closer to 🌖🇺🇸
>
> — **NASA Administrator Jared Isaacman** @NASAAdmin on X · [November 14, 2026](https://x.com/NASAAdmin/status/2121715782350504106)

## The booster that turned away

Booster 22 was supposed to fly back to the tower about six minutes after liftoff. It lit 32 of its 33 Raptor engines for the boostback burn that turns it back toward Texas, and the burn ended early, according to Payload. Kate Tice of SpaceX said the engines had stopped firing too soon. The booster’s guidance system sensed it was out of position, steered it toward open water and set off its self-destruct system, [The A.P. reported](https://www.nbcnews.com/science/space/spacex-launches-15th-starship-flight-aborts-booster-recovery-maneuver-rcna609274). Buoys marked off the wreckage in the Gulf.

SpaceX’s explanation changed during the flight. On the webcast, Mr. Huot first said the booster had called off the catch after detecting a guidance and navigation problem, [The Next Web reported](https://thenextweb.com/news/spacex-starship-flight-15-booster-catch-aborted). He later said guidance wasn’t the cause, and Ms. Tice said early indications pointed to a “low fuel mass alarm,” [according to Space.com](https://www.space.com/space-exploration/launches-spacecraft/spacex-says-starship-flight-15-booster-aborted-chopsticks-landing-catch-after-launch-due-to-guidance-issue-video). SpaceX hasn’t named a root cause. It was not clear whether the Federal Aviation Administration would require a mishap investigation.

On Hacker News, where a thread on the catch drew close to 900 comments, one reader saw the abort as a matter of margins.

> Had they put another couple of tonnes of propellant and oxidizer on the Booster they might have caught both of them.
>
> — ChuckMcM on [Hacker News](https://news.ycombinator.com/item?id=50502003)

The alarm Ms. Tice described fits that reading. But SpaceX hasn’t said whether the booster actually ran short or an instrument misread, so the theory is still a reader’s guess. The bigger point is how little the newest booster has shown. SpaceX caught Version 1 boosters on Flights 5, 7 and 8, in October 2024, January 2025 and March 2025. Four Version 3 boosters have flown since May. Two were lost in high-speed impacts with the water, one made a controlled splashdown and Friday’s was destroyed offshore, according to the [public flight log](https://en.wikipedia.org/wiki/List_of_Starship_launches).

Reflights have been slower still. Booster 14, caught in January 2025, flew again 131 days later on a flight that ended in failure. Booster 15 waited 221 days before its second launch, in October 2025. Neither was caught the second time.

## The heat shield

Mr. Musk set the bar for the upper stage himself. In November 2024 he wrote on X what he considered the hardest technical problem Starship had left.

> The biggest technology challenge remaining for Starship is a fully & immediately reusable heat shield.
>
> Being able to land the ship, refill propellant & launch right away with no refurbishment or laborious inspection. That is the acid test.
>
> — **Elon Musk** @elonmusk on X · [November 20, 2024](https://x.com/elonmusk/status/1859068810692644949)

About 18,000 ceramic tiles cover the ship’s belly. The shields on Flights 10 and 11, in 2025, came back patchy and uneven. Flight 13 on July 24 was the second flight of the Version 3 ship, and imagery afterward showed only a handful of tiles missing. Mr. Musk then said he considered the heat shield problem solved, [Tech Times reported](https://www.techtimes.com/articles/322079/20260729/starship-heat-shield-tiles-are-dead-end-rapid-reuse-former-nasa-experts-warn.htm).

Three people with long NASA careers looked at the same pictures and disagreed. Dan Rasky spent nearly four decades on heat-shield materials at NASA’s Ames Research Center and helped invent the material in the Crew Dragon capsule’s shield. He wrote an analysis with Charles Camarda, a former astronaut, and Charles Miller, who led the Trump administration’s NASA transition team. They pointed to white streaks at the boundaries between tiles, a possible sign that hot gas was getting into the seams, and to cracked tiles and broken edges, [Gizmodo reported](https://gizmodo.com/starships-heat-shield-looked-great-after-flight-13-but-experts-spotted-something-troubling-2000791792). Mr. Rasky told Ars Technica:

> Starship’s current thermal protection system is a dead-end for all missions that require full and rapid reusability
>
> — Dan Rasky, former NASA heat-shield researcher at Ames Research Center, [gizmodo.com](https://gizmodo.com/starships-heat-shield-looked-great-after-flight-13-but-experts-spotted-something-troubling-2000791792)

The claim is narrower than it sounds. The three called the Version 3 tiles an incremental step past the shuttle’s, built on the same basic approach. They said the tiles would still let Starship be reused, with long turnarounds, and judged that good enough to launch Starlink satellites. Their doubt was about reflying within hours, which is SpaceX’s target. The company can’t yet turn a ship around in days or weeks.

SpaceX kept changing the design. For Flight 14 it added tile retention in the areas most at risk, closed newly found paths for plasma to get behind the shield, tried curved tiles and flew two tiles recovered from an earlier ship. Sven Bilén, an engineering professor who studies reentry heat shields, laid out those changes in an essay for The Conversation, [republished by Phys.org](https://phys.org/news/2026-09-starship-orbital-flight-shield.html). Two tiles, he wrote, can’t show how many flights a whole shield will last.

While Ship 42 was in orbit, SpaceX photographed its shield with long-range ground cameras and with cameras on some of the Starlink satellites it had just released. Mr. Huot said the heat-shield team judged the shield ready for reentry. Afterward the company said every system on the ship, the shield included, appeared to have worked well. The A.P. reported that it was not clear whether SpaceX would fly the captured ship again.

In his September essay, Dr. Bilén wrote that a recovery would not settle the matter by itself.

> For me, the next telling milestone will be recovering a Starship and then inspecting it and preparing it to launch again.
>
> — Sven Bilén, engineering professor who studies reentry heat shields, [phys.org](https://phys.org/news/2026-09-starship-orbital-flight-shield.html)

## The shuttle’s record

NASA pitched the space shuttle to Congress as a way to “make spaceflight cheap and frequent.” Roger Pielke Jr., a science policy researcher at the University of Colorado Boulder, said in 2011 that “it was originally expected to fly into space about once per week” and that “Its actual flight rate was one-tenth of that,” [NBC News reported](https://www.nbcnews.com/id/wbna42543397). Planners had said an orbiter could be ready to launch again in as little as two weeks.

The fleet flew 135 missions over 30 years, an average of 4.5 a year, with a peak of nine in 1985, according to [a history of the program’s shortfalls](https://en.wikipedia.org/wiki/Criticism_of_the_Space_Shuttle_program). Readying an orbiter between flights usually took months. Before the Challenger accident, about half of that time went to unplanned tests and modifications. The record turnaround was 54 days, between two flights of Atlantis.

Writing in *Nature*, Mr. Pielke and Radford Byerly, also of the University of Colorado, put the average cost at about $1.2 billion a launch in 2010 dollars across 131 missions from 1982 to 2010. Counting lifetime program costs, the figure rose to $1.5 billion. NASA had spent more than $192 billion on the fleet by the end of 2010.

On Hacker News, the comment that drew the most replies made the comparison in Starship’s favor.

> Starship is, essentially, Space Shuttle done right: fully reusable booster without solid boosters (avoids Challenger disaster) and orbiter stacked on top instead of on the side (avoid Columbia disaster). ... They can build a booster ever 4 weeks and a ship every 2 weeks.
>
> — GMoromisato on [Hacker News](https://news.ycombinator.com/item?id=50502268)

The design points are fair as far as they go. Starship has no solid rocket motors, and its ship rides on top of the booster. The production rates the reader cited could not be independently verified. Fast production also lowers the cost of a new rocket, not the cost of readying a used one. Another reader replied that the shuttle’s compromises came from serving NASA and the Pentagon at once, and from spreading work across congressional districts to protect its budget.

A third reader objected on safety.

> The Space Shuttle took 2 people into orbit on its first launch and returned them safely to earth. ... Starship did not and has not done that and until it does I will continue to hold reservations.
>
> — mk\_stjames on [Hacker News](https://news.ycombinator.com/item?id=50502366)

The history is accurate, and it cuts both ways. The shuttle carried crews from its first flight and still never came near its turnaround goal, and two orbiters were lost with seven astronauts aboard each. Starship has carried no one. SpaceX’s case for it rests on the step the shuttle never managed: getting a vehicle back from orbit and ready to fly again in days rather than months.

## Turnaround

SpaceX’s timetable is short. On the company’s third-quarter earnings call on Oct. 20, [covered by rews](https://rews.cc/a/on-an-earnings-call-in-texas-spacex-promises-to-catch-both-h-4006bb), Mr. Musk said “at first, there will be a little bit of refurbishment that’s needed between each flight.” The goal, he said, is that “you can immediately land the rocket and launch it again just like an airplane,” according to [a transcript of the call](https://www.fool.com/earnings/call-transcripts/2026/10/20/spacex-spcx-q3-2026-earnings-call-transcript/). Gwynne Shotwell, SpaceX’s president and chief operating officer, said the company aimed to “refly the second stage by early next year.” Mr. Musk said he expected weekly Starship flights at some point in the first half of 2027 and “twice weekly or maybe even more frequently than that by the end of next year.”

Falcon 9 is the company’s own precedent. One Falcon 9 booster, B1067, flew its 37th mission on Aug. 25, two short of the 39 flights of the shuttle Discovery, [Tech Times reported](https://www.techtimes.com/articles/325493/20260825/falcon-9-workhorse-nears-shuttle-reuse-record-cape-canaveral-starlink-era-ends.htm). That record covers the first stage alone. Mr. Musk said on the call that a double catch would be “the first time that an orbital launch vehicle has been fully recovered.”

![A line of satellites with orange solar panels recedes into a starry navy sky above the curved turquoise Earth.](https://rews.cc/img/2bcf9db0b75a09a5ad18cf3bcd45851152b54ae5.webp)

Weekly flights also need a regulator’s approval. The FAA’s May 2025 decision allows 25 Starship launches and 50 landings a year at Starbase. In October the agency began a full environmental impact statement on a SpaceX plan for up to 100 launches a year in a first phase and 456 launches and 912 landings a year in a second, [SpaceQ reported](https://spaceq.ca/starship-launches-456-a-year-starbase-faa-review/). The FAA’s notice said rates would rise “as the vehicle continues to mature and reliability improves.”

## Refueling in orbit

The moon landing needs a test no one has flown. A Starship lander can’t carry enough propellant from the ground to reach the moon, so SpaceX plans to fill it in Earth orbit from tanker Starships. NASA’s inspector general reported that the first ship-to-ship transfer demonstration, once planned for March 2025, had slipped to March 2026, according to [a summary of the program](https://en.wikipedia.org/wiki/Starship_Propellant_Transfer_Demonstration). It slipped again. “We’ll attempt an on-orbit refueling early next year,” Ms. Shotwell said in October.

How many tanker flights a landing needs is in open dispute. On Oct. 22, Charlie Precourt, a member of NASA’s Aerospace Safety Advisory Panel, gave an updated estimate at the panel’s public meeting, and Mr. Berger posted it.

> Just now, on the Aerospace Safety Advisory Panel call, member Charlie Precourt said the latest estimate for refuelings needed to support a Starship landing on the Moon is “around 22” launches.
>
> — **Eric Berger** @SciGuySpace on X · [October 22, 2026](https://x.com/SciGuySpace/status/2113340439893856278)

Mr. Musk [replied on X](https://x.com/elonmusk/status/2113377866478231647) that the figure was “way too high” and that the real number would be “more like 6 to 8.” Mr. Precourt’s estimate allowed for operational reserves and for propellant that evaporates in orbit, [New Space Economy reported](https://newspaceeconomy.ca/2026/11/09/how-many-orbital-refills-will-spacex-starship-need-to-land-astronauts-on-the-moon/). In 2021 the Government Accountability Office found that SpaceX’s original proposal called for 16 launches: a depot, 14 tankers and the lander. In November 2023, Lakiesha Hawkins of NASA said a landing would need launches “in the high teens.”

The count matters because each tanker is a full Starship launch. Heat leaking through the steel tank walls boils off propellant while the depot waits for the next tanker, and measuring that loss is one purpose of the demonstration. Under NASA’s revised plan, Artemis III will stay in Earth orbit so its four astronauts can practice docking with Starship and Blue Origin’s Blue Moon lander, The A.P. reported. Artemis IV, aimed for as early as 2028, would be the first crewed moon landing since Apollo.

![A cream watering can receives orange liquid from a floating bottle and sprinkles orange droplets beneath a crescent moon against a blue background.](https://rews.cc/img/f6adf7f926beccd786df74bf141e003495006c95.webp)

## The cost argument

Mr. Musk has put numbers on Starship for years. In November 2019 he put the cost of a mission to SpaceX at about $2 million. In February 2022 he said a launch would cost under $10 million “all in” within two to three years, according to [a tally by New Space Economy](https://newspaceeconomy.ca/2026/06/21/how-accurate-were-elon-musks-falcon-9-and-starship-kg-claims/). Both figures were internal costs, not prices, and SpaceX had published no Starship price as of June.

Falcon 9 shows how far apart cost and price can sit. A 2018 NASA paper by Harry W. Jones put the shuttle at $54,500 a kilogram to low Earth orbit and Falcon 9’s advertised price at $2,720, about a twentyfold cut, according to [the paper on NASA’s technical server](https://ntrs.nasa.gov/citations/20200001093). Falcon 9’s posted price has risen anyway: about $54 million in 2011, $62 million in 2018 and about $74 million for a dedicated launch in early 2026, by New Space Economy’s count.

SpaceX’s own filings hedge. Its prospectus for this year’s public listing said for the first time that Starship doesn’t need full reuse to launch the next generation of Starlink satellites, [TechCrunch reported in May](https://techcrunch.com/2026/05/26/starships-path-to-reusability-looks-murky-after-spacexs-s-1/). Tim Farrar, a satellite industry analyst, wrote in a note to clients:

> If this reusability is not achieved then the cost of launch on Starship may not be much lower than Falcon 9
>
> — Tim Farrar, satellite industry analyst, [techcrunch.com](https://techcrunch.com/2026/05/26/starships-path-to-reusability-looks-murky-after-spacexs-s-1/)

Without reuse, he estimated, a launch could cost as much as $100 million, or $1,000 a kilogram. That figure holds even if Starship reaches its full 100-ton capacity, which he called “by no means a foregone conclusion.” On the October call, Mr. Musk said Flight 14 flew about half loaded and still delivered Starlink capacity equal to 10 Falcon 9 launches.

The optimistic case has an economist behind it too. Karen Jones, a space economist at the Aerospace Corporation, modeled three scenarios in [a paper on super heavy-lift rockets](https://csps.aerospace.org/sites/default/files/2026-06/Jones_SuperHeavyLift_FINAL20260614.pdf). Two assume a launch first costs $100 million and fall to $133 to $233 a kilogram for a fully loaded Starship after 10 reuses. The most optimistic starts at $50 million and reaches $67 a kilogram after nine. Every one of them assumes the vehicle actually flies that many times.

On Hacker News, one reader questioned demand rather than cost.

> Falcon9 already made it a lot cheaper but i don’t see much market opening up? other then launching their starlink’s
>
> — emkoemko on [Hacker News](https://news.ycombinator.com/item?id=50502614)

SpaceX’s own figures lend the reader’s point some weight. Of the company’s 35 launches in the third quarter, 21 were Falcon 9 Starlink missions, and both Starship flights carried Starlink satellites, SpaceX said on its call. Starlink brought in $11.4 billion last year, according to the prospectus, and Mr. Musk has called Starship essential to holding down Starlink’s costs. A cheaper Starship pays off for SpaceX first. Whether outside customers see lower prices is a separate question, and the record doesn’t answer it yet.

[Writing in The Space Review](https://www.thespacereview.com/article/5350/1) on Nov. 2, before the flight, Burak Oktenli, a Georgetown graduate student who studies launch resilience, argued that a launch provider can cut its own costs without passing all of the savings on to buyers. He also wrote that frequent refurbishment can erode the advantage.

> A beautiful catch is a milestone in recovery, but alone is not proof of affordable reuse.
>
> — Burak Oktenli, writing in The Space Review, [thespacereview.com](https://www.thespacereview.com/article/5350/1)

## What comes next

Flight 16 is aimed at December and could be the first Starship launch from Launch Complex 39A at the Kennedy Space Center, [NASASpaceFlight reported in October](https://www.nasaspaceflight.com/2026/10/spacex-flight-15-16-milestones/). FAA planning material lists Dec. 10 as a possible date, which SpaceX hasn’t confirmed, [Tech-Insider reported](https://tech-insider.org/spacex-starship-florida-launch-date-2026/). Earlier this year the FAA approved up to 44 Starship launches a year from that pad.

Ms. Shotwell’s target for flying a recovered upper stage again is early 2027. As of Saturday, SpaceX hadn’t said whether that ship would be Ship 42.

The next scheduled step is local. The FAA holds public meetings on the plan for up to 456 Starbase launches a year in Brownsville on Nov. 17, online on Nov. 18 and in Port Isabel on Nov. 19. It is taking written comments under docket FAA-2026-12808 through Nov. 30.

## Join the discussion

- [SpaceX catches the Starship to end Flight 15](https://news.ycombinator.com/item?id=50500963) — Hacker News · 1,316 points · 898 comments

## See also

- [SpaceX third-quarter 2026 webcast transcript](https://s21.q4cdn.com/184289198/files/content_files/SpaceX-Third-Quarter-2026-Webcast-Transcript.pdf) — s21.q4cdn.com · The company's edited transcript of the Oct. 20 call on reuse, cadence and refueling
- [Space Shuttle Operations and Infrastructure, NASA technical paper, 2005](https://ntrs.nasa.gov/api/citations/20050172128/downloads/20050172128.pdf) — ntrs.nasa.gov · NASA's own account of what it took to turn an orbiter around between flights
- [Starship HLS](https://en.wikipedia.org/wiki/Starship_HLS) — en.wikipedia.org · History and schedule of the Starship moon lander and its refueling plan
- [From 25 to 456 launches a year: the FAA's Starbase review](https://spacelawjournal.com/starship-environmental-review-faa-starbase-peis/) — spacelawjournal.com · How the environmental review works and what the higher launch cap would allow
- [What's next for SpaceX's Starship after Flight 15](https://www.space.com/space-exploration/launches-spacecraft/whats-next-for-spacexs-starship-after-epic-flight-15-chopsticks-catch) — space.com · Space.com's look at the flights and milestones that follow the catch
- [With Starship Flight 10, SpaceX prioritized resilience over perfection](https://techcrunch.com/2025/08/27/with-starship-flight-10-spacex-prioritized-resilience-over-perfection) — techcrunch.com · The 2025 flight where SpaceX deliberately stressed the heat shield
- [Starship flight 15](https://en.wikipedia.org/wiki/Starship_flight_15) — en.wikipedia.org · Reference page on the flight's vehicles, plan and timeline

## Sources

- [Starship flight 15 powers to orbit—setting up a possible catch on Saturday](https://arstechnica.com/space/2026/11/starship-flight-15-powers-to-orbit-setting-up-a-possible-catch-on-saturday/) — Ars Technica
- [SpaceX returns Starship to its launch tower—but there's a catch](https://arstechnica.com/space/2026/11/with-starship-catch-spacex-still-shows-it-can-dazzle-the-world/) — Ars Technica
- [SpaceX catches Starship with launchpad arms in dramatic first](https://aljazeera.com/news/2026/11/15/spacexs-starship-captured-by-launchpad-arms-in-dramatic-first) — aljazeera.com
- [SpaceX aborts Starship booster catch, but the ship is still in orbit](https://thenextweb.com/news/spacex-starship-flight-15-booster-catch-aborted) — thenextweb.com
