Elon’s Thursday Starship Presentation

What Eric Berger is hoping to learn.

[Thursday update]

It’s at 2100 EST, on Youtube.


[Update a few minutes later]

Bob Zimmerman will be live streaming it as well.

[Friday-morning update]

Here is Eric Berger’s report.

What I heard last night that was new to me (maybe he’s said this before) was that it was less than a million per flight on the margin, and it could deliver 150 tons to LEO (I had thought the number was a hundred tons). That’s a fifty percent increase, and a one-third reduction in cost per pound.

[Late-afternoon update]

A report from NASA Spaceflight.


[Saturday-morning update]

More on Musk at Politico, in which Yours Truly is quoted and cited.

72 thoughts on “Elon’s Thursday Starship Presentation”

  1. Berger is the only reason I look at Ars Techica (otherwise known as the Beth Mole Covid Propaganda Newsfeed). I sometimes wonder, but he’s better than Jeff Faust (sic).

      1. Not like it was 3 years ago, though. The entire site is noticeably worse. And that’s not counting the commenters, who are a horrible mess.

    1. Reading Berger’s post-presentation article, I can only reiterate: if he wants to be taken seriously as a space journalist, he needs to find a job somewhere else. People don’t just read his articles, they read the comments to it, and he is manifestly writing for an audience of Woke morons.

    2. Besides Berger, there can be decent stuff on physics, chemistry, some astronomy, some computer reporting and archaeology. Anything that’s relatively apolitical and isn’t just a reprint from Vice is liable to be reasonable. But stay away from Covid, health policy, energy policy, EVs and anything climate related. Total Loony Tune stuff and the commenters on those topics literally foam at the mouth.

      1. Even the supposedly apolitical topic commenters have distemper now. Know any knitters? Ask them about Ravelry. Specific to Ars Technica, archaeology now ends up on climate change. And you’ll have to throw in reprints from Wired alongside Vice.

          1. It’s an interesting question. Should we read the parts of Ars we like, even though a click for Eric Berger is tacit support for Beth Mole? Isn’t that how we lost the Culture Wars and got where we are now? When are supporters of cancel culture canceled themselves, whether we like them or not?

          2. “Should we read the parts of Ars we like, even though a click for Eric Berger is tacit support for Beth Mole?”

            Ad blockers, baby. I won’t give money to sites where a significant portion of the commenters want me dead.

  2. There are so many potentially exciting thing that Elon could give in his update. e.g.:
    – FAA expectation and options
    – Timeline for further development milestones
    – Details on the catching tower
    – Orbital refueling & depots
    – Images of interior design
    – Candidly addressing SLS (not gonna happen)
    – Accepting refundable reservations for private transport!

  3. The long-term life support question reminded me of a suggestion I made in the 80s-

    On long duration missions to Mars or elsewhere, the food stored for the crew would eventually be turned into propellants. By running all human waste products through a supercritical water oxidizer, the outputs are CO2, water, nitrogen and a tiny amount of minerals. The CO2 and part of the water can be converted via a Sabatier reactor to methane and oxygen and simply stored in the propellant tanks, while the nitrogen is used for atmosphere makeup against leakage. The minerals can be stored and added to the Mars city biome, so the mass of the food is not parasitic but is turned into an asset.

  4. No one asked about Starship cargo doors and Elon volunteered no information about them either. Ditto for detailed information on what still remains in the works on “Stage Zero” at Starbase. So it goes.

    Unsurprisingly, the main unknown is the nature of the upcoming FAA environmental assessment. A FONSI or FONSI-with-mitigations ruling means we could see a first orbital launch in March or April. If a full EIR is mandated, focus shifts from Starbase to LC-39A and that first orbital launch attempt happens in FL in late 4Q 2022 or, more likely, sometime in 2023.

    In any case, the LC-39A Starship launch facility should be done around year’s end and the same could be true of one of the ex-petroleum platforms. I’m guessing one probable result of an FAA-mandated EIR for Starbase would be an acceleration of both efforts and very probably the refitting of the second ex-oil rig in rough parallel with the other and with the land-based facility at LC-39A.

    I have speculated for some time that an abort capability for Starship would likely consist of chilling in its engines on the pad so that they could push Starship free of a disintegrating Super Heavy even on the pad. With most versions of Starship likely to get three more Rvacs apiece, the needed thrust would certainly be available. It was nice to see this approach being confirmed by Elon.

    I infer, based on Elon’s remarks about Raptor 2.0, that SpaceX is going full-bore with production of all Raptor 2.0 components except the still insufficiently durable combustion chambers. Apparently, the plan is to put the pedal to the metal on combustion chamber production once a suitable “final” design is in-hand and to then quickly complete the partly-built engines.

    1. I think the tower and GSE must be about done, since he said, permission notwithstanding, they could be ready to launch by mid-March. I think the first version of payload bay doors will be the sliding panel shown for Lunar Starship, which would also be a good idea for Mars Starship. An interim Starlink rotary dispenser could be made to work with that as well. I don’t know that the “chomper” door shown early on is practical, even in zero-G. Remember, we did see a nosecone where they cut Shuttle-style doors, a while back, but not since. I wonder if the long refilling rail shown will determine the size and location of whatever type doors they build.

      1. The big GSE stuff is done, but the daily summary videos from NSF still show a lot of smaller stuff being brought in and installed. How much of that is left to do is not obvious by inspection.

        There is also at least a modest amount of work remaining to be done on the launch mount and launch tower. My rough gauge for completeness of such work is the amount of scaffolding still in place.

        The LN2 and LOX portion of the GSE seems to be at least minimally complete and functional, but the LCH4 portion is not. I have seen it said that TX law requires an extra interior fence around any large LNG storage facility. We have seen some revision of previous tank farm infrastructure and the construction of such a fence begun recently. Let’s hope this satisfies the TX tank wallahs and soon allows WDRs and static fires to begin.

        Cargo door design seems to be a work-in-progress including multiple prototype exercises. More than one of these may well see tests on-orbit before serial production commences.

        I think the “guide rails”are actually the piping and cable raceways on the dorsal sides of Starships, though they also may be equipped to serve as alignment aids for refilling. That would be a very SpaceX thing to do. But I think these rails/raceways stop short of the lower edge of any cargo door.

  5. The biggest point in the presentation is Plan B: Build a Starship launch tower at KSC LC39A and turn Phobos and/or Deimos into a landing platform with a “catch tower.” That’s a slight condensation of Musk’s actual words, including the phrase “catch tower.” He also said he wanted to have Plan B ready to go by the end of 2022.

    The thing I found most interesting, other than Plan B, was the discussion of head film cooling. I’d like to see the techical details on that, since I’m really only familiar with throat film cooling (introduced in the V2 more than 80 years ago). He also noted they might not have orbital “refilling” perfected for a year or two (so, by the end of 2023?).

    1. Thinking about it, I wonder about the pattern of the oxygen-rich and fuel-rich gas injectors. I bet that pattern is a trade secret. I wonder if the fuel injectors are on the periphery, and the oxygen ones in the center, or if they’re intermingled. I assume the film coolant is methane, which would ignite of injected next to a hot oxygen gas stream, which seems like it would defeat the purpose. Me too dumb to figure it out.

        1. I don’t think Raptor has an injector plate in the combustion chamber, though it may have one in each pre-burner. Or perhaps the pre-burners, like Merlin, use pintle injectors. In any event, no liquid propellants are delivered to the combustion chamber, just two streams of already-combusting gas – one oxygen-rich and the other fuel-rich. Which also explains Elon’s reference to film cooling in the combustion chamber as well as in the nozzle throat.

    2. My take on that “refilling by the end of 2023” thing was that it was said in context of having to implement Plan B if the FAA comes in wanting a full EIS for Starbase. That would push the entire schedule rightward by six months to a year.

      But I could be wrong about that. The whole presentation had an air of last-minute improvisation about it. The use of obviously old CGI footage for the refilling section of the presentation suggests either insufficient time to gin up new footage more in keeping with current plans or, perhaps, the state of flux of those plans. There was no propellant depot ship shown, for example, and none of the questions received noted this omission. Perhaps that is because the design of the depot ship has yet to be nailed down sufficiently to turn it over to the CGI crew.

      If testing can actually start in March or April – as I think likely – we will see EDL, then cargo doors and Starlink deployments as initial highest priorities. This means refilling experiments would be late-2022 or early-2023 activities.

      The necessity of implementing the Plan B tilt to KSC would, of course, push refilling to the right by the same amount as everything else on today’s tentative schedule. Right now, the Starship development schedule is Schrodinger’s Cat. Once the FAA hands down a finding anent Starbase, the quantum uncertainty will disappear, one way or the other.

  6. Since side by side refilling precludes the end to end ullage thrust transfer method I wonder what “pumping” method is to be used now?

    1. Starship, like all spacecraft, has six degrees of freedom: pitch, yaw, roll, and X, Y, Z translation. So ullage thrust transfer will work on any axis. (In this instance, the tanker goes “up” and the receiver goes “down” and the fuel/oxydizer flows from the tanker to the receiver.)

      1. Doh! Yes, I see that now. And by altering between say +Z and -Z they could maintain a relatively stable orbit during the process.

        1. It seems that a slow, continuous, thruster-defined “barrel roll” might be best for achieving the needed artificial gravity differential while minimizing orbital altitude variance. As a side benefit, it would also keep the thermal gradient over the entire two-ship complex minimized whether in sunlight or in shadow.

          1. Wouldn’t that push the fuel and oxydizer to the “outside” wall of the roll? Translation away from target tankage (so the fluids would flow “downhill”) would be low acceleration, maybe only enough to maintain the orbit of the accumulation tanker (phrase used so SpaceX doesn’t have to use the verboten “depot” in reach of tender legislative ears).

          2. I was assuming the thrusters would keep the receiving Starship on the outer locus of the roll and the tanker on the inner locus.

            Given that continuous low thrust from the sides of both Starships and from the dorsal side of the receiving Starship would be needed to execute such a slow barrel roll, the paired Starships would probably have to be canted a bit relative to the radius of the roll so the resultants would put the tanker’s sump at the “lowest” point in the artificial gravity field anent the tanker – which would also be the “highest” point in said field anent the receiving Starship.

            Thrusters on the dorsal – i.e., heat shielded or “windward” – side of an EDL-capable receiving Starship would, one assumes, be hidden beneath small heat-shielded doors. On a depot ship with no need to ever do an EDL, the thruster installations on its dorsal side would be much more straightforward.

  7. Btw, Rand, the original presentation from 4 years ago suggested it wouuld cost $50mln to fly 100 people to Mars (whence the $500K tickets). This was one passanger ship and 6 tankers, so an average launch cost of a little more than $7mln. Now it’s under $2mln on the margin, maybe $10mln all up (that last a Musk guess from last night).

    The 100 ton to LEO payload was always sandbagged, and has been seized on by various haters to come up with huge numbers of tanker flights per mission, with maximum cost inflation to show it’s just as bad as SLS, somehow. The current figure of 150 tons is to a “useful” orbit (such as Starlink or ISS). But the refilling orbit is a minimum orbit, which allows each tanker to bring up 200+ tons. We don’t know how much plus because the design isn’t finalized. The Raptors keep getting better (more efficient), the ship and booster lighter. Musk has suggested Lunar Starship may only require four tankers.

    1. He has mentioned a nine-engine stretched 220 ton to orbit second generation Starship recently too. He hinted at it the other night.

    2. Assuming the HLS Starship retains its notional 1,200-tonne full propellant load, then four tankers would imply 300 tonnes per tanker mission. That might just be possible as Raptor continues to add muscle and the whole stack continues to slim down, but will probably not be an initial operational capability.

      That would also likely require the LEO refilling orbit to be about the same as that currently used for Starlink deployment. That means the propellant depot ships would have to use modest amounts of propellant to keep themselves in orbit long-term.

      This would also have an advantage anent the design of the LEO propellant depot ships in that they might be able to sport only minimal insulation layers and no active boil-off reduction/elimination systems. Instead, the boil-off could be calibrated to exactly match the propellant mass required to maintain them in their low orbits.

      Any propellant depot ships detailed for service in cis-lunar space would, obviously, have to be both better insulated and equipped with active anti-boil-off equipment.

      1. The new Starship standard tankage is 1500 tons. The reason Lunar Starship might only need 4x 200 ton refillings is because he thinks LSS doesn’t need to be any bigger than that. I wonder if he’s visualizing a short-length LSS?

        1. Could be. There is certainly no obvious advantage to be had in stretching the HLS Starship design to match the new, longer, “standard” Starship dimensions. Given the usefulness of those “panniers” between the Rvacs on the current design of the HLS in terms of both center of gravity and readiness-to-hand of part of the cargo payload, I don’t think HLS will get the 6-Rvac treatment either. A longer and up-engined EDL-capable Starship makes sense for many posited mission profiles, especially for tankers, but not for lunar landers.

  8. I’ve seen talk of SpaceX moving operations if they get caught up in a protracted series of studies but couldn’t they lobby congress to pass legislation that would give them an exemption? They could hit up the Space Force to help with the lobbying since SpaceX is one of our most important strategic assets.

    1. SpaceX wouldn’t move operations entirely to FL, it would simply shift emphasis to FL until the EIS for Starbase was completed. There seems no credible reason to believe such a study would come in with a mandate for completely shuttering Starbase.

      The effective impact to the overall program schedule would be limited to however long it takes to build out the LC-39A Starship facility to the same degree as Starbase has been as of now. Building a second such complex should go appreciably more quickly based on the experience gained in building the first. So we’re looking at a matter of months, not years, even if the EIS for Starbase somehow manages to take years to complete.

      As such, I think SpaceX has no need to try strong-arming the government for anything. That would simply provide gratuitous ammunition for his detractors in and out of government. If the FAA finding is for a new EIS, then Musk would be best served by simply dialing back Starbase a bit, cranking the FL Starship effort up to 11 and providing a bit of salutary shock and awe to his enemies by continuing the roll of the Starship juggernaut barely impeded.

      1. Musk knows before long he’s going to need many launch and landing sites. He’s got LC39A (one Falcon pad, Starship pad soon to come). If I were him, I’d be eying LC39B, and eventually converting LC-40 and 41, 49 down the road, and 36 after Blue goes belly up. Eventually the Space Coast will be the SpaceX Coast, and old Missile Row will be the new Starship Row.

        1. The new Starship complex at LC-39B is already a-building and two more such at LC-49 are likely to be started within a year. One or both of the current pair of ex-oil rigs will be coming on-line about then too. LC-39B might become available in as little as two or three more years and I can certainly see SpaceX snapping it up. SpaceX might eventually convert SLC-40 to Starship ops, but I think it will still have value as a Falcon 9 pad for some time. SLC-41 could fall to SpaceX within five years as I don’t really see a path forward for ULA once the next round of NSSL contracts is decided. SLC-36 would only be available if Blue completely withdraws from the launch business. That’s possible but not, in my view, too likely. Four or five Starship pads on land at KSC/Canaveral would be plenty if a steady stream of ex-oil rigs, acquired for scrap prices, is processed through Pascagoula and Brownsville then moored offshore of both Starbase and the Space Coast.

      2. I don’t disagree but they should look at all the options available to them. SpaceX is a key national security resource and should be treated as such, and there is likely support in congress should it be focused to help out. Lobbying isn’t the same as strong arming.

  9. 1 million tons to Mars.
    10,000 trips
    $10 million times 10,000 is $100 billion.
    How long before it could happen?
    Is there a faster way?

        1. Lunar market could exceed 50 billion dollars within 10 years or
          be 5 to 10% of global space/satellite market.

          1. Now, for less obvious point.
            Could you send people to Mars, from the lunar surface?

            Or, can send people to Mars, faster {3 months or less], from a higher Earth orbit.

            And more crazy, can mine Mars water and ship it to Venus orbit, and go from Venus to Mars.

          2. Using the Moon depends on whether there is is minable lunar water.
            Musk doesn’t have mine lunar water.
            But since NASA is paying for Starship, you could count
            that payment as being paid to explore the Moon.
            If lunar water is mineable is He3 mineable?
            If He3 is mineable, then one get a lot of H2 and He4 and other stuff, like water.
            So, if He-3 is minable, what Musk wants is H2 and CO2- he doesn’t have to toss out the Helium {He-3 and He-4}.
            A factor of whether lunar water is mineable is if lunar CO2 is mineable. Or if got 10% water and 5% CO2, it make the 10% water more mineable. Or if got 5% water and 10% CO2, it makes the 5% water more mineable.
            Or if lunar water is mineable and it’s if related how much CO2 is minable. And if He-3 is mineable, Musk can be concerned with getting CO2 and H2, for sending people from the Moon to Mars.
            Or Musk could mineable lunar water, and have company worth say 10 billion within less than 5 year, and sell that company, and focus mining H2 and CO2 as relates to going to Mars.

          3. The Moon doesn’t require water to operate Starships, of whatever types, in cis-lunar space. As Elon noted several times during Thursday’s presentation, Starships need a lot more oxygen than methane mass to operate. Oxygen will be readily available on the Moon from smelting of regolith into metals – no water necessarily required.

          4. “The Moon doesn’t require water to operate Starships, of whatever types, in cis-lunar space. As Elon noted several times during Thursday’s presentation,”

            Water is 8/9th oxygen.
            If the Moon doesn’t have mineable water, Musk shouldn’t try to Mine Lunar H2.
            No one should bother with the Moon.
            If Moon has mineable water, lunar water will be mined. Musk could mine lunar water, and then sell the company which can mine lunar water. Or keep it.
            Once one starts to get Mars settlements, then you can mine the Moon- for Hydrogen and Iron {and other stuff] and then Moon can export water. And then you have lunar settlements- when water imported to the Moon is about $1 per kg. Before water cost $1 per kg on the Moon, or even when $1000 per kg, one can have lunar hotels {lunar tourists} and/or lunar bases.

          5. [[and then Moon can export water. ]]
            Meant, “and then Moon can import water. ”

            Eventually, one put lunar space elevator, import water, get electrical power from dropping water on the Moon.
            And lunar water will be cheaper than $1 per kg, and electrical power will be cheap.
            Or lunar electrical should become cheaper the electrical power on Earth, but with space elevator, it will be even cheaper.

    1. $2mln x 10,000 is only $20 billion. Musk’s Visa probably has a higher limit than that.

      My question is, how many people does that million tons on Mars represent? I bet 100,000.

    1. Mars needs to be explore to find mineable Mars water.
      Meanwhile, a Mars base can use LH2 and cement. NASA can ISRU, sand and gravel. NASA could probably manage mining Mars sky for the CO2. Hydrogen and CO2 makes water. {And methane}.

      1. Or just tell me where can one mine on Mars, 1000 ton of Mars water for 1 million dollars {$1 per kg}.
        A Mars settlement need about 1 million tons of water for $1 billion dollars.
        And a settlement site which within region, within 10 km radius, with 1 billion tons of water which can mine for $1 per kg/ $1000 per ton [or much less]. Or 1 billion tons mine water, delivered to people in a town for less than 1 trillion dollars over time period of decades.
        US uses .6 trillion tons of water per year, or 6 trillion tons per decade. And roughly cost about $.1 per cubic meter. And significant portion is used by electrical powerplants.
        Or Mars needs water for cheaper electrical power generation.
        And that waste heat, can used to warm stuff.

        [And I think Mars should export water to Venus. Using mass drivers.]

      2. Huge quantities of Martian ice have already been located via orbital observation. Accessing it will simply be a matter of landing in a place with subsurface ice and doing a bit of drilling. There is no need to “make” water on Mars.

        1. Drilling for water, seems like the cheapest way to get water on Mars.
          On Earth when drill for water, you get different kinds of water.
          On Mars it seems one also get different types of water.
          And seems possible the water one could get is salt water or water which polluted in some way.
          It seems if going to live somewhere on Mars, an important aspect related to real estate value, is the quantity and quality of available water.

          1. It’s expensive building towns which then become ghost towns, because the people could move to somewhere of better real estate value.

          2. We just wont know what the regional water/ice conditions will be until someone prospects different locations. The same is true for processing and how to access.

            Before we get to providing water to a settlement, we need to do the little things like prospecting and pilot plants testing different methods. How would this initial step be taken?

            People always talk about the business case for space and talk about these mature systems and then the conversation breaks down because there are some missing steps. The business case for space is how can a company make money right now and when looking at it that way, the business case is providing parts or services to existing companies with a strategic eye to the future.

            I think the same is true for water on Mars. Long before we get to providing water for a colony, there needs to be a chain of products and services that grows to facilitate the existence of water works.

  10. That Politico article: OUCH!!!

    ““They are shitting the bed,” said a top Washington space lobbyist who works for SpaceX’s competitors and asked for anonymity to avoid upsetting his clients.”

    1. Loss of voluntary sphincter control is not uncommon among people who see their inevitable doom approaching at flank speed.

  11. A few years ago there was an old-space lobbyist who used to write anti-SpaceX screeds while posing as an “industry expert”. What was his name? I wonder if he is the “potty-mouthed D.C. lobbyist” mentioned in the Politico article.

  12. ““Starship can carry a C-17 load of cargo and get it anywhere in the world in an hour,” said Walker, who previously served on a SpaceX advisory board but no longer has any ties to the company.”

    So you are saying that the Talibans would be flying Superheavies and Starships right now and be the leading country in space activities?

    1. That’s kind of apples and oranges, because the cost per pound on the expendable will surely be a lot higher. I’d rather buy two 150s than one 250 for the money. I don’t have any 250T payloads.

      1. For Moon or LEO space station where don’t want to return second stage [you want use it- for whatever on lunar surface- or part of space station in LEO]

  13. If lift Starship empty, you should able to bring first stage to orbit. Yes?
    And nicer if can make starship taller, instead 120 meters, it taller. Maybe need to two of them for stick artificial gravity station.

  14. Interesting news that SpaceX is giving Dragon EVA capability and is developing EVA suits. Jared Isaacman will demonstrate both next fall on his Polaris Dawn mission. He also means to be aboard the first crewed Starship launch.

Comments are closed.