LC 31: Flight 13 Flies, Booster 20 Doesn't Come Home

MaxOkay, welcome back to Landing the Moon. Okay, so get this. Starship Flight 13 finally flew on July 24th.

BlakeThird attempt, right?

MaxThird attempt. And it deployed satellites this time. 20 Starlink V3s. Starship's first operational payload drop ever.

BlakeThat's interesting. So the ship worked.

MaxThe ship worked great, Booster 20 not so much.

BlakeWait, what happened?

MaxNailed the boost back burn,

BlakeSo

Maxlooked flawless, then lost the landing burn and slammed into the Gulf.

Blakesplit result. One win, one crater.

MaxBasically, yeah.

BlakeWe're spending real time on that landing burn today because the engine count everyone's citing doesn't line up.

MaxOh, you're going to love this part. We're also getting into Ship 40's single Raptor relight in space. Tiny maneuver, huge implications for Artemis.

BlakeArtemis—and I want to press on whether the FAA treats this the same way it treated the fifty two day grounding back in May.

MaxPlus we're grading our own LC-3 predictions against what actually happened with those oxidizer turbo pumps.

BlakeAnd closing with something falsifiable: does SpaceX even attempt a tower catch on Flight 14.

MaxSo let's start where it went sideways: Booster 20, the Gulf, and a landing burn nobody saw failing.

BlakeWalk me through what we actually know.

MaxSo the scorecard out of Starbase this week is one for two, and the one is wild.

BlakeOne for two! Walk me through the one.

MaxJuly Twenty third, five fifty one p m Central, Ship Forty lights all thirty three Raptor Three engines on Booster Twenty and actually gets off the pad.

BlakeThird try, right?

MaxThird try. There was the T-0 mission abort on the sixteenth—engines wouldn't light—then a weather scrub on the twenty third.

BlakePersistent at least.

MaxAnd then get this: Ship Forty deploys twenty Starlink V3 satellites on a suborbital path.

BlakeWow.

MaxFirst time Starship has ever released an operational payload. CNN called the flight "Lucky Thirteen."

BlakeNot a mass simulator, not a banana, an actual working satellite stack.

MaxTwenty of them designed to reenter, which they did.

BlakeWhat happened to the ship after that?

MaxSo check this out: Ship Forty splashes down in the Indian Ocean. tips over like it usually does and just keeps transmitting.

BlakeAfter it tips over?

MaxNASASpaceFlight reported the ship stayed intact and kept sending data after it rolled, which let SpaceX pull imagery of the tiles and engines post-splashdown.

BlakeUseful test data, whether or not that was the plan.

MaxSure, the plan.

BlakeNo, the Booster.

MaxRight, so Booster Twenty nails the high-thrust part of the boost back burn. all thirty-three engines textbook.

BlakeAnd then?

MaxThen it loses the landing burn, hits the Gulf hard. No soft splashdown this time.

BlakeSo a flawless upper stage deploy and a booster that face-planted into the water.

MaxFace-planted is generous, but yeah.

BlakeI want to separate what got demonstrated from what got claimed in the press release. Payload deploy real, booster loss also real. Both happen on the same flight.

MaxLook at you refusing to pick a side!

BlakeI'm asking which one actually moves the needle for Artemis.

MaxThat's a whole different argument, because buried in this flight is a demo that barely made headlines: a single engine relight in space.

BlakeThe one they skipped on Flight Twelve.

MaxThat one. Okay, so buried in that whole splashdown story, Ship 40 coasted about 27 minutes, then relit a single Raptor engine in space.

BlakeThe demo flight 12 skipped.

MaxExactly. That relight failed on Flight 12 after the vacuum engine shut down early. This time, one engine, one relight, clean.

BlakeOne engine out of how many on the ship?

MaxSix on Ship 40. But it's not about the count. It's about proving the plumbing works cold-soaked in vacuum. That's the whole ballgame for landing on the Moon.

BlakeWalk me through why engine turns back on doesn't sound like a lunar milestone to me.

MaxBecause Artemis III needs Starship to land on the Moon, then take off again. No relight, no ascent. Every trip to the surface assumes the engines restart after sitting in vacuum for days.

BlakeSo this is step one of a long staircase.

MaxThree steps removed minimum. Tech Times quoted Kent Chojnacki, NASA's Deputy Manager for the Human Landing System Program. His line, version 3 is what HLS is going to be based on.

BlakeWait, this vehicle? The one that just lost a Booster in the Gulf?

MaxThis one.

BlakeWow.

MaxSame Raptor, same relight architecture NASA's banking the Lunar Lander on.

BlakeOkay, then let's do the arithmetic because that's where I get uneasy. What does the refueling chain actually look like?

MaxEstimates floating around put it at 10 to 16 tanker flights for one lunar mission just to fill the depot before HLS heads to the moon.

BlakeSixteen launches for one landing?

MaxCan Space Policy Online flag the real culprit on the high end, cryogenic boil-off, propellant warming and venting off of the depot faster than SpaceX can top it off?

BlakeSo if you need over a dozen flights bunched close together, what cadence does that require? Has anyone shown SpaceX can launch that fast back-to-back without a scrub?

MaxNot yet; not close. This flight alone took three attempts just to get off the pad.

BlakeRight; so one relight in a test flight is not the same as proving a fleet can refuel a depot on schedule.

MaxCorrect; it's necessary; it's nowhere near sufficient.

BlakeAnd the booster, the thing that just missed its landing burn, that's asking the same restart question just thirty miles up and minutes earlier.

MaxSame physics, different consequence. Ship 40 gets a scorecard. Booster 20 gets a crater.

BlakeWhich is exactly the tension I want to sit with going into the failure itself.

MaxShooting back to that relight talk, now the landing burn, because the numbers don't even agree!

BlakeWait, what do you mean they don't agree?

MaxWikipedia's account has ten of thirteen engines relighting for the landing burn, and two of those shut down within seconds.

Blake10 of 13.

MaxRight, but Spacecom's live coverage says it looked more like five engines still firing right at its splashdown.

BlakeThat's a real gap between 5 and 10.

MaxYeah, and nobody's reconciled it. Both outlets were watching. We're watching the same feed!

BlakeSo we don't actually know the engine count; we know the water hit hard either way.

MaxDude, hard is an understatement. Ten engines, maybe five, either way you're not landing softly with half your relights.

BlakeOkay, walk me back to May Twenty Seventh, because that's where this actually starts.

MaxRight, Flight Twelve's booster loss. The mishap report closed July Thirteenth—fifty two days grounded.

BlakeAnd what did it find?

MaxHeat effects on engine hardware plus bad alarm thresholds that were tripping engines that didn't need to trip.

BlakeSo SpaceX flew four corrective actions on Booster Twenty for exactly that problem.

MaxAnd boost back worked—clean separation. Duration—clean burn back toward the coast.

BlakeBut the landing burn didn't; so did the fixes work or not?

MaxHonestly, different question. The failure moved down stream, it didn't disappear.

BlakeThat's the regulatory piece I want pressed; same booster, lost twice in two flights.

MaxAnd a totally different response.

BlakeWalk me through that gap. Flight Twelve got a full FAA mishap investigation.

MaxNASA Space Flight reports the FAA isn't requiring one for flight

Speaker 3fourteen.

MaxFlight Thirteen

BlakeSo losing a booster on ascent gets fifty two days grounded but losing it on landing gets a shrug.

MaxRight, that's basically the read, yeah.

BlakeWhat's the actual distinction, public safety versus hardware performance?

MaxThat's the split: ascent failures threaten people on the ground; landing failures mostly threaten SpaceX's own hardware.

BlakeDry distinction, if you're the one building the rocket.

MaxFair, but it means Booster twenty one isn't sitting on a regulatory

Speaker 3desk.

MaxRepertory clock this time.

BlakeWhich changes the whole cadence fight we were having two episodes back.

MaxOh, we're absolutely circling back to that abort call. With the FAA question parked, let's settle the other one. Wikipedia's write-up on the July 16th abort points to six Raptors whose oxidizer turbopumps never spun up to flight speed.

BlakeSix, not the twenty-nine of thirty-three they got quoted right after the scrub.

MaxSame sources, same day, but six turbopumps failing to spool up before liftoff is pretty specific.

BlakeSo let's go back to LC-30. You said that a board prove the flight software works.

MaxHere we go.

BlakeI mean what's that claim resting on? The computer caught a fault before ignition, sure, but that's the fourth time we're talking about engines that won't start clean.

MaxFine, I'll grade myself; I was right the abort logic did its job; I was wrong calling it isolated; six turbopumps failing together is a pattern.

BlakeMy call was the mirror image: repeated start failures as the real signal mostly holds up.

MaxMostly?

BlakeI didn't expect them flying the full stack again eight days later with a clean liftoff. The fix worked at least once.

MaxOnce; so what actually proves it's fixed?

BlakeThree straight flights, ascent, booster relight, ship relight with zero engine start anomalies anywhere in that chain.

MaxI'd add this: same turbopump supplier, same design, still causing trouble on flight fourteen, then it wasn't fixed!

BlakeIt was covered up.

MaxRight, and one good flight doesn't clear anybody, SpaceX, the FAA or us for calling it too early.

BlakeSpeak for yourself.

MaxFair, but the schedule doesn't grade on effort, it grades on dates.

BlakeAnd NASA's got a whole new set of dates of its own now, after February's restructuring.

MaxBuilding on that restructuring flag, let's talk about what CSIS actually laid out in February.

BlakeOh boy, hit me. CSIS reports NASA turned Artemis 3 into a low-earth orbit lander and spacesuit demo, targeting mid-2027. The actual crewed moon landing? That's Artemis 4, no earlier than 2028.

MaxNo earlier than is doing a lot of lifting in that sense. sense.

BlakeSame restructuring canceled the Exploration Upper Stage all right, paused Gateway after roughly $4.4 billion already under contract, and floated a $20 billion lunar surface base spread over seven years.

MaxWait, they paused Gateway after spending $4.4 billion on it?

BlakeThat's the number, yeah.

MaxCool, cool, cool.

BlakeAnd Astronomy.com flagged something sharper: if Starship keeps slipping... NASA could lean on Blue Origin's Blue Moon Mark II for that twenty twenty eight attempt.

MaxA second lander provider. Everyone's framing that as a fix.

BlakeYou don't buy it?

MaxI don't buy it as a fix. It's a second schedule to miss, not an insurance against the first one.

BlakeWalk me through that.

MaxBlue Moon Mark II hasn't flown crew, hasn't done an uncrewed landing, nothing. You're not de risking Artemis. You're doubling the number of unproven vehicles NASA's. Mass's betting the mission on.'

BlakeCNBC put Starship spending north of fifteen billion dollars already; that's the sunk cost pulling against switching horses.

MaxRight-so nobody actually walks away from Starship-they just add Blue Origin on top, and call it redundancy.

BlakeOr call it a second delay with better PR So put a stake in the ground. Artemis IV, crewed landing. Your date?

MaxI'll say twenty thirty and I'll be an optimist saying it.

BlakeI'm not touching a year yet, but Flight 14 is coming fast, and that one's got its own reckoning.

MaxOkay, closing ritual, dated calls, no hedging; Flight 14 to SpaceX go for a tower catch.

BlakeI'll go first, no catch. Ship 41 splashes down again. Booster 21 attempts the tower by September 15th.

MaxI'll top that full catch attempt on both stages flying before Labor Day, September seventh.

BlakeWalk me through why the FAA moves that fast after grounding them- them twice already.

MaxFlight 13 didn't trigger another 52-day pause. That's the tell. The agency's loosening up. Plus, NASASpaceFlight.com flagged Ship 40's clean flight as raising catch potential for the next mission.

BlakeI read that too, but I'm not sold. My call bets the opposite. The FAA opens a formal review into Booster 20's landing burn and it drags into October before Flight 14 even gets a date. The date.

MaxSo next month we're grading two completely different timelines.

BlakeOn the record, both of us.

MaxAnd if Flight 14 slips past October, I'm officially renaming it 14th Times the Charm.

BlakeDon't jinx it.

MaxToo late. See you next month for the scorecard. Okay, before we let you go, Flight 14 predictions, Dated, on record.

BlakeSo we can grade ourselves properly this time.

MaxLike Elsie thirty, yeah-all right, I say SpaceX attempts the tower catch on Booster Twenty One by September fifteenth, and they pull it off.

BlakeHmm-bold-I'll take the other side of that. My call: the FAA doesn't close out the Booster Twenty landing burn investigation before October First. That relight problem is quieter than the abort, but it's not simple.

MaxWill hold each other to it."

BlakeWe will.

MaxAnd that Ship Forty relight still the one number that actually matters for Artemis Four.

BlakeAgreed; everything else is noise.

Maxuntil that's repeatable.

BlakeIf this helped you follow Artemis, send it to someone still confused about why we're going back to the Moon.

MaxSubscribe and we'll see you back here to check the math.

BlakeLanding the Moon out.