Is Today The Day?

I’m assuming they have clearance from the FAA now.

[Update a while later]

Here’s the story on what happened with the FAA. I’d like to understand more detail on it. I’d be surprised if SpaceX really knowingly launched without authorization. I wonder if there was a miscommunication?

[Update a few minutes later]

Bob Zimmerman has more, with links.

[Post-flight update]

[Thursday-morning update]

You have to appreciate that kind of honesty in a CEO.


[Afternoon update]

Per the comment that he was being sarcastic:

69 thoughts on “Is Today The Day?”

    1. Mr. Biden doesn’t cast a shadow. He is a holographic projection — just like Robert Picardo on STNG.

          1. They solved faster-than-light travel and communications, teleportation, force fields, and can rapidly cure any ailment… but not baldness.

          2. Ishmael, I’m bald. I know it isn’t a fashion statement. My body just produces too much dihydrotestosterone.

  1. Most of the space news organizations are operated by known leftists; they report more space news than the lamestream media but they’re no more fair or balanced. Keith Cowing may be the most irritating, but his work is of poor quality and he’s a person of no significance. Space News and Jeff Foust are more prominent leftists in the space business. If anything, I suspect there was fine print misunderstanding and the Biden administration is using it to make an example.

    Rand, you got a couple mentions in the comments section of Foust’s latest woke hit piece.

      1. Not really any context:
        SocketHeadCapScrew • 18 days ago
        Space is full of right wing nut jobs. Rand Simberg is the posterchild.

        Which generated some comments, one by Richard Malcolm:
        Rand’s a libertarian gadfly, not a right wing nut job.

        Then a general disintegration:
        DP Huntsman:
        Yeah, afraid that’s correct. He’s been completely in the anti-fact, anti-evidence, anti-truth, right wing media bubble these last 4 years- and he will NOT challenge that crap he’s emerged his head in. It’s all he does.
        Richard Seaton:
        Simberg is the very definition of nut job. He and Trump could have been soulmates.
        Nothing said by the Trumpist’s here belongs anywhere but in a sewage tank.

        Rand, can you elaborate on what it’s like to emerge your head in crap? LOL
        Really ArsTechnica territory there.

          1. Sorry, you’ll have to ask someone who’s actually done that…

            Well not head actually, but waist arms and elbows pulling an eel yes. Post septic system, it’s actually blackish/gray and dries as gray in color. Pungent yes, but not merely as much as you might suspect. Unlike for instance, rotting soybean or decomposing dog carcass, the odor does not induce vomiting.

        1. “Richard Seaton” is of course longtime combox nutjob and libertarian loather Gary Church under his latest sockpuppet account. Probably the only thing every single person in the SpaceNews comboxes (or indeed any other comboxes now that Paul Spudis’s blog is shut down) can agree on is that Gary is an insufferable troll.

          Huntsman is just another TDS sufferer. Just so we’re clear, since I’m the very same Richard Malcolm who commented: I am quite fond of libertarian gadflies like Rand. 🙂

      1. LOL Take it up with Dr. Foust! Rand, I forget exactly – and going back it looks like the comments devolved/ballooned to the point SN shut down the comments. Something about you and Bob Zimmerman being right wing nutjobs.

  2. “Prior to the Starship SN8 test launch in December 2020, SpaceX sought a waiver to exceed the maximum public risk allowed by federal safety regulations,” an FAA spokesperson said Tuesday. “After the FAA denied the request, SpaceX proceeded with the flight.”

    I have a very hard time accepting this masterpiece of weasel worded, Monday armchair quarterbacking. This almost borders on defamatory. This is disgusting coming from the FAA. Totally vague, could mean anything. What was the specific items sought to be waived? The FAA doesn’t say. SpaceX will be very wise to remain silent about this but at the same time they need to put out an immediate PR stating that they continue to adhere diligently to all aspects and requirements of the FAA licensing regime. Put the ball back in the FAA’s court to say what it was done wrong. Vague isn’t trustworthy. The job of the FAA is far too serious and important to be playing these games. FAA you f*ked up on 737Max IMO the onus is on you. DO YOUR FUCKING JOB!!!

    Rand you wrote a book about this. How the hell does one mitigate public risk??

  3. Somone dropped a comment offering a possible explanation in over at the main SpaceX subreddit just now, and I have to say, it has a plausible ring of truth to it:

    As a former auditor who has quite a lot of experience with regulators: this is pretty carefully worded to ensure that both parties can walk away without looking bad, whilst making sure everyone knows that the problem was at the SpaceX end. It’s a high quality piece of beaurocramancy in that regard.

    The precise nature of the non-compliance is unclear, but the fact that the FAA has cleared SN9 for testing very quickly would certainly push me to think that it was largely about miscommunication. SpaceX probably looked at the conditions and felt they were within the agreed limits, the FAA disagreed.


    1. Until something goes wrong with SN9 and then the rug gets pulled?

      Love letters between FAA and SpaceX don’t cut it. Sorry it’s the engineer in me coming through. You either meet criteria or you don’t. If you do and something goes wrong, you re-visit the criteria, but you DO NOT do it retroactively. You admit you missed something and then move forward. This is what adults do.

    1. That’s what it looked like, yeah. One engine started, and it actually went past vertical and was trying to correct when it ran out of sky.

  4. Yeah stop motion of the Spacex feed shows one engine lit and the other trying to relight but failing.

    The camera was at the landing spot looking up….very neat.

    Ok well – learn from it and then move on to SN10.

    1. Groundhog Day

      They get to do this over and over and over again until the finally get it right.

      Actress Andy McDowell fits into this someway or another, but I haven’t figured it out yet.

  5. Eric Berger: “S9 NEIN.”

    How different this feels from an SLS test setback. We know SpaceX got loads of data, and they’ve already got SN10 sitting out there on the other launch mount, just waiting for its engines to make another hop in a few weeks – with yet more prototypes at various stages of completion over in the high and mid bays. Move fast, build lots of hardware, break things, figure it out.

    By the time the first one of these things does its landing burn on Mars, they’re gonna know every failure mode backwards and forwards, with live operational data to back it up.

    1. The trouble is, the engines have frequently not lit up during static fires. The landing mode requires 100% engine restart reliability, and they aren’t even near that for ground starts.

      Also, during the relight, it looks like some parts flew off from somewhere in or around the engine bay.

      1. The Raptor is not a finished product yet, obviously.

        But yes, it’s something thay have to master to make this work.

        If Starship/SH were fully expendable, this would all be far easier. But the whole pont here is to break away from the expendable paradigm.

  6. So what is the escape mode for passengers when one or both engines don’t fire at the flip at 2000 feet. Maybe they should initially restart all three engines in case one doesn’t start to guarantee a controlled landing.

    1. I would think everyone would have zero-zero ejection seats?

      In some prior thread on SN8 I think I posted some calculations about keeping Starship horizontal for landing and using sets of Super Dracos for braking, for both Earth and lunar touchdowns. But they didn’t take that design path, and are committed to making the current method reliable, as the simplicity, weight, and cost will be lower than any alternative.

    2. I’ve been thinking, if only to comfort future passengers, they should try an artificial soft deck hover. The problem with that is the need for excess fuel and the subsequent creation of a flight profile they don’t intend to use.

      Still, I’d like to see a few landings before I’d ever get near that thing. Then again, I listened to the latest Mike Rowe podcast and agree with him that astronauts have a different perspective of risk than I do.

  7. That bellyflop looked worse than SN8. I’ll need to go back and look at SN8, because I don’t recall that much venting during the flight. I was also concerned by what looked like a fire raging in the aft compartment after MECO. I wonder if the vented O2 got sucked into the aft compartment and something that wasn’t meant to be burned got burned.

    1. I would expect each bully flop test to look worse as they explore the flight envelop. This time they had a pretty steep nose-down orientation and recovered from it just fine.

      I would think nose-down is the worst case for a Starship to be in because the fuel is going to be below the engines and all the flow would have to run uphill. If vapor is going to enter the propellant feed system, that’s likely the scenario that would cause it.

  8. Looks like the engine bells may have hit each other in the video around 11.48- after engine 2 flamed out. There must be a lot of turbulence across those bells on that restart. That has to be a complicating factor. But if the bells hit each other, it’s over.

  9. RE most likely there being less to the FAA problems than meets the eye, I’ll repost this from Disqus yesterday:

    RE the FAA delay, my advice is, don’t panic. At least not yet.

    The average bureaucrat’s average reply to anything new is “no, fill out all the paperwork again and we’ll think about it.” It’s universally understood in the trade to be the safe way to reach retirement with full pension intact.

    The only exception is if the bureaucrat STRONGLY understands that top agency political management WANTS them to say “yes”, and will take the heat if there’s then a problem. This was I gather the case for FAA regarding space launch flight experiments under the previous two White Houses, maybe even a bit longer.

    Now there’s a new Administration. There is no need whatsoever to invoke any desire to “get Elon” to explain what we’ve seen so far. Just working-level bureaucrats who have not heard whether the implicit “go” of the last decade-plus continues under the new top management, and are thus suddenly dotting i’s and crossing t’s they knew they could skip a month ago.

    In other words, so far the facts are easily explainable by the FAA’s new management simply not having given the matter any explicit attention either way. So my advice is, don’t panic, wait and watch.

    Meanwhile, I’ll repeat something I have been saying for years: It is a HUGE mistake to drag contemporary political fractures into space issues. ANYTIME you do so, you automatically turn one half or the other of your potential supporters into reflexive opponents. Space issues tend to cut diagonally across the standard political divides, and are one of the few things that can bring together people from both sides and inspire them to work together productively. Making this out to be some political thing without absolute proof can only turn it into one – to the vast detriment of our collective future. Please don’t do it.

  10. As for the actual test flight today, I’d say A: The fact that it happened bears out the first half of my previous post, and B: SpaceX has some work to do on getting Raptors to relight reliably in flight.

    They also now have a whole lot more data on what the relight problem may be, so I expect it’ll be solved soon. So they can move on to the NEXT surprise, at a very non-traditionally high rate of progress.

  11. One simple change, which would only carry a small fuel penalty, is to set up to execute the flip turn higher up, so you could still flip using just one engine, while starting the alternate engine for decel and touchdown. Or you could try to light all three and then idle the redundant one, or run all three at a lower throttle setting until you need even less thrust for touchdown.

    So, some quick engine-out statistics, assuming that the only failure mode is engine failure. Here S is the engine reliability rate (hopefully 90+%) and F is the engine failure rate per start attempt. All engines are assumed to be good prior to the restart attempt.

    Trying to light a given two engines gives four possible outcomes: both lit, only one lit (two ways), and none lit.

    The landing only succeeds if both are lit, and the odds of that are S^2, or (1 – F)^2. The landing fails at a rate of 2F – F^2. As the failure rate gets small, the crash rate stays at about twice the engine failure rate, as you would expect.

    In the three engine mode, there are eight possible outcomes. All engines light, two of three light (which happens three ways), one of three light (which happens three ways), and none light.

    In the three engine case, the landing fails at a rate of 3F^2 – 2F^3. The ratio of the two failure rates is an uglier equation, so I’ll use a quick table of engine failure mode versus flights between RUD

    engine reliability, engine failure rate, flights per 2 engine RUD, flights per 3 engine RUD


    Rel, fail, T for 2, T for 3.
    90%, 10%, 5, 36
    95%, 5%, 10, 138
    96%, 4%, 13, 214
    99%, 1%, 50, 3356
    99.5%, 0.5%, 100, 13378
    99.75%, 0.25%, 200, 53422
    99.9%, 0.1%, 500, 333556

    In the 3 engine relight scenario, 96% restart reliability crosses the 1 in 200 mission failure mark, whereas in the current 2 engine relight, 1 in 200 failures requires 99.75% restart reliability.

    Although the three engine method has a small risk of failure in the early attempts, as the engines still go through development and teething pains, it would probably never see a landing failure due to engine failure, even in thousands and thousands of flights. The two engine method likely will see a landing failure even if the engines become extremely reliable.

    1. “One simple change, which would only carry a small fuel penalty, is to set up to execute the flip turn higher up, so you could still flip using just one engine, while starting the alternate engine for decel and touchdown. Or you could try to light all three and then idle the redundant one, or run all three at a lower throttle setting until you need even less thrust for touchdown.”

      Why not plan to use one engine higher, and after flip, start 1 and/or 2.
      And if the 1 engine does not start, then start the other 2

  12. It appears that the large off-vertical angle at impact reduced the local loads on the landing pad and did less damage to it than the upright SN8 crash. The flight control system might have that consideration programmed in, and if it learns a few seconds before impact that impact will not be survivable, it could sacrifice itself for the greater good.

  13. You have to appreciate that kind of honesty in a CEO.

    I interpret that as sarcasm. I think Musk is saying in effect “You think we didn’t consider that?”.

    1. Jim, that’s a possible case, but I’m of the opinion that their engine thrust calculations assumed a much better re-light success rate than we’ve seen in reality. At this point if they don’t have all three available next time around I’d be surprised.

      I am also curious why they are waiting until the last second to do the flip – seems like something that could be done more conservatively until they get it working. If it’s that critical then there’s a design issue.

    2. Honestly I think NASA could learn from Elon’s general bluntness in regards to failure. Part of the reason it catches so much heat for failures is because it’s management often tries to convince the public and Congress that everything is completely under control. Then when something blows up everyone wonders why. While the SpaceX method is to say “We expect the first few attempts to fail. We’ll just keep changing the design until it no longer explodes “

      1. Which NASA would do as well: Except their test cycle would be 2-3 years between launches, using all space-rated hardware.

    3. They could have considered it, done a risk analysis, and determined two was the right number. They have had quite a bit of experience in relighting Merlins. Success can breed its own institutional blind spots.

    4. The thing about Elon is, you can easily see him meaning it either way.

      Anyway, I guess his Twitter sabbatical didn’t last very long.

    5. I read it as more “We considered it and chose not to do it for reasons that seemed perfectly… reasonable… at the time. Clearly, we chose… poorly.”

      1. Not stupid so much as untimely.

        Criticism of a NASA test can sit in the can for weeks, months, or years but when writing about SpaceX, you have to be as fast as they are.

  14. Latest from the acting head of NASA:

    “I know what the timelines are for the [Space Launch System], but it’s hard to determine what the timeline is and capabilities are for the Starship,” Jurczyk said in a new interview with Futurism.

    What I know about SLS timelines is that they continue to move in the wrong direction.

    1. Ironically, Jurczyk is probably more right thank he knows, but for the wrong reasons.

      It’s easy to know the timelines are for SLS: Always in the future, in a “free beer tomorrow” sort of way.

      It’s harder to know the timeline for Starship because:
      a) Elon is a bit over-enthusiastic and aggressive when it comes to his timeframes, so they slip
      b) the old guard has no faith in Elon because he’s testing and failing publicly
      c) when Elon’s timelines slip, they rarely only slip a small amount, rather than at an increasing rate the way SLS timelines do
      d) SpaceX is doing things much differently than if he were running a cost-plus workfare program.

      Even now that Resilience successfully reached and docked at ISS, I’m sure there are still people saying that they don’t think SpaceX knows what they’re doing.

    2. Capabilities should be easier to plan around than timelines. How long would it take NASA to design a payload that would utilize Starship/Super Heavy? Might as well start now.

  15. I don’t want to rain on Elon’s parade, but once Space X starts testing the Superheavy booster, they can’t have so many explosive events, even with the offshore platforms. One, it’ll be more expensive. Two, the bangs will be a lot louder with a 230 ft booster blowing up. Three, too many launch failures would provide an excuse for the Green busybodies and their friends in the FAA to interfere, and possibly even stop entirely, Elon’s plans for space colonization. So, he may have to take a little bit more time between test flights with the Superheavy. Finally, how much of a beating can the Phobos and Deimos platforms take after suffering through multiple Superheavy crashes? Is Musk going to reinforce them structurally?

    1. they can’t have so many explosive events
      A contrarian would point out people like explosions. As long as no-one gets injured it’s one of the all-time favorite collective human experiences, second only to fireworks. If a rich guy says “I like explosions, and nobodies gonna get hurt”, how would you address him?

      1. One would have to be crazy to think that the testing was designed to increase the risk of a boom in order to entertain the nerds who watch hundreds of hours of live streams where nothing happens. Even crazier to think that testing engineers dramatic events so that the finale of a landing has a big phycological payoff for the audience that has been watching livestreams for years where most of the action are birds flying into frame.

        SpaceX doesn’t go out of their way to insure that their test articles go boom but they certainly aren’t letting the possibility stop them from pushing capabilities and testing limitations. When something happens regardless of what you choose to do, the logical course is to take advantage of it rather than worry about the small number of people who wont like it.

        It is cool for now but soon enough it will be back to boring and only obsessives will be watching.

          1. Cool guys don’t look at explosions they caused, they are always looking at the audience for the reaction. =)

    2. Stands to reason that the work they are doing now will prevent what you are worrying about. There will be some additional challenges with Super Heavy and the platforms but most everything will be ironed out with the Starship testing. And will they be testing Super Heavy at their current launch facilities? Construction of Super Heavy prototypes are further along than the platforms, which look untouched.

      1. With exactly twice the takeoff thrust of a Saturn V, my bet is on the Superheavy launch. I was three miles away from the Skylab Workshop launch (last Saturn V), and it was so loud that I couldn’t shout loudly enough for the guy next to me to hear.

  16. As much as I try to dislike that guy he’s pretty cool and he can launch rockets. He’ll be buried on Mars – assuming he doesn’t invent a way to live forever.

Comments are closed.