Showing posts with label BFS. Show all posts
Showing posts with label BFS. Show all posts

Sunday, June 13, 2021

Starship to take over Starlink launches

 Here, I talked about how Starlink and Starship were linked: Starlink would provide the cash to help build Starship and Starship would be so big it could launch the thousands of satellites needed to get the Starlink's satellite constellation working.  Prophetic!  Remember, you read it here first (2017).

Now, Musk has announced that Starship will take over Starlink launches, replacing Falcon 9 in this rĂ´le.

Since dedicated Starlink launches began in May 2019, Musk, COO and President Gwynne Shotwell, and a few other SpaceX officials and executives have made it clear that the company would ultimately transition the task of launching and maintaining the Starlink constellation from Falcon 9 to Starship. Barring major surprises, Starship is being designed to be fully and rapidly reusable from the ground up, nominally making the system far cheaper to launch.

After Musk announced a radical redesign that replaced carbon composite structures with simple steel, Starship may even be far cheaper to build than Falcon 9 or Falcon Heavy – despite being several times larger, heavier, more powerful, and more capable. Despite its relative shortcomings, though, Falcon 9 has become an extraordinarily reliable and available workhorse for SpaceX and has completed 28 operational Starlink launches – delivering ~1670 satellites to orbit – since November 2019.

However, while Falcon 9 has done and continues to do an extraordinary job of routinely launching satellites and astronauts, Starship promises to blow it out of the water. It might be several years before Starship is deemed safe and reliable enough to launch humans but SpaceX could feasibly start launching Starlink satellites on the rocket almost as soon as it begins orbital flight tests.

Thanks to the low cost of each Starlink satellite, likely now around ~$250,000, it would be surprising if SpaceX didn’t include at least a few dozen satellites in the early phases of orbital Starship flight tests – even if success is far from guaranteed. At some point, though, and perhaps quite quickly, Starship will safely make it to orbit, reenter, and touch down beside a Super Heavy booster a few times in a row, effectively demonstrating fitness to launch (uncrewed) payloads.

It could take a bit more proof to convince paying customers with satellites worth tens to hundreds of millions of dollars to entrust launch contracts to Starship but SpaceX itself – likely to be the builder and owner of the world’s largest satellite constellation for the indefinite future – has more flexibility to tailor its appetite for caution. With the capabilities Starship could feasibly offer, SpaceX might also be hard-pressed to just sit and wait.

[Read more here]

The first test flight of Starship and Super Heavy will prob'ly take place in July, though SpaceX won't try to land either vessel for re-use--both will splash down into the sea.  Could the first launch of a few Starlink satellites on Starship happen before the end of this year?  At $2 million per launch, Starship will be just 4% of the cost of a Falcon 9 launch, but will be able to launch 400 satellites at a time instead of just 60, cutting the effective launch cost of each satellite by 2 orders of magnitude.

Starlink will provide a truly global high-speed internet service, and it will change the world as well as making SpaceX the sort of cash needed to build an inter-planetary space fleet.  

The plunging cost of lifting a kilogram to orbit won't just change space.  It will change the Earth too.


Source: Neopork

Wednesday, September 16, 2020

Musk: next stop 60,000 feet

 From Inverse


The Starship, SpaceX's vessel designed to one day take humans to Mars, is set to fly higher than ever soon. Over the weekend, SpaceX CEO Elon Musk outlined plans for an upcoming prototype model of the ship, dubbed "SN8," to fly up to 60,000 feet (or around 20 kilometers) into the air and return to Earth.

The launch would be the highest any model of the Starship has ever flown, beating this month's "SN6" launch, last month's "SN5" launch, and August 2019's "Starhopper" launch. All these ships flew to a height of around 500 feet (or 150 meters). The flight would also send the ship soaring to nearly double the altitude of regular commercial airplanes, which cruise at a comparatively low height of around 35,000 feet.

The feat could pave the way for some of SpaceX's grandest plans. The company is currently developing the ship at its Boca Chica facility in Texas, with the goal of a fully-reusable rocket capable of transporting over 150 tons or 100 people into space at a time. The ship could take on satellite launches similar to the Falcon 9, while its liquid oxygen and methane fuel enables passengers to explore Mars and return by refueling with the planet's resources. Musk has previously outlined his goals to build a city on Mars by 2050.

On Saturday, Musk explained this next phase of the rocket's development via his Twitter page. In around one week, SpaceX expects to complete an "SN8" prototype. This new model will sport flaps and a nosecone, looking closer to the final concept images than the simple silo that flew last month.From there, the team plans to hold a static test fire. This is a routine part of SpaceX's launches where the engines are tested before an actual launch. Then there will be checkouts, then another static fire, then the flight itself will be held.



This will be a big step forward,  There will likely be failures along the way, but if this works, the next step will be orbital flight, where the ceramic heat shield is tested.  Since these prototypes are 1/30th the cost of carbon-fibre composites, SpaceX can afford to 'waste' them with repeated testing.  SN8 will be the first to test the 'sydiving' technique.  But parts for SN11 have already been seen.  Progress is fast.


Sunday, June 14, 2020

SpaceX's Mars timetable

Starships at Mars Base Alpha


Elon Musk has confirmed the 2022 cargo and (late) 2024 crewed expedition dates to Mars.

From the Orlando Sentinel:

Falcon 9 is a means to an end, for Elon Musk, and while the rocket that just helped transport astronauts to the International Space Station celebrated its 10-year anniversary on Thursday, Musk is keeping his eyes on the prize with Starship.

Musk took to Twitter to reiterate the company’s plans for the next-generation space vehicle and its timeline to get to Mars.

Responding to questions on the social media platform, Musk confirmed SpaceX plans to get Starship on an uncrewed mission to the fourth planet from the sun by 2022 and with humans by 2024.

“Starship is the key to making life multiplanetary & protecting the light of consciousness,” Musk wrote.

He also figures that it will take about 25 years, or roughly 12 transfer windows - when Mars and Earth are in the right position to make the trip economical - for whatever presence is on Mars to be self-sustainable, not relying on support from Earth. 

Musk has previously stated his belief that the human species needs to branch out and colonize on other planets.
“We’re faced with a choice. Which future do you want? Do you want the future where we become a space-faring civilization and are in many worlds and are out there among the stars or are forever confined to Earth? And I say it’s the first, and I hope you agree with me," Musk said in a speech last September to SpaceX employees at the company’s testing site in Boca Chica, Texas.

“The reality is as far as we know this is the only place at least in this part of the galaxy, or in the Milky Way[as a whole], where there is consciousness and it’s taken a long time for us to get to this point,” Musk said.

On Wednesday, in response to a dire tweet about the eventual demise millions of years from now of the Earth at the hands of the sun, Musk tweeted, “That’s when all life on Earth will be boiled off. What matters is how long civilization is capable of making the jump to Mars. This could be a very short period of time measured in decades. It took 4.5 billion years to get to this point & civilization isn’t looking super stable.”

This has been my estimate of the timetable for over a year now, with minor tweaks:






  • H1 2020: Tests to 20+ kms altitude. [Now delayed till July 2020]
  • End 2020: Full stack (i.e., Super Heavy booster plus Starship) operational.  First orbital flights of the Starship.
  • Early 2021: First commercial customers (for satellites), launches of Starlink constellation
  • Late 2022: Uncrewed mission to Mars (Mars is in opposition in December)
  • 2023??First commercial space station.  Launched on Starship, built by non-SpaceX companies—or maybe even by SpaceX  [Not part of SpaceX's plans]
  • 2023: 'Dear Moon' circumlunar expedition
  • 2024: Moon Base Alpha [Depends on NASA]
  • Late 2024/Early 2025: Crewed mission to Mars  (How many ships?  Here's my analysis)
  • Early 2027: Second expedition to Mars.  Return of at least one Starship.  
  • 2029: Third expedition to Mars.  Martian population reaches 300.  If Starship works, SpaceX will start designing even bigger rockets, capable of carrying more than 100 passengers, so Mars's population in 2029 could be more than 300.  Return of some Starships.
  • Monday, June 1, 2020

    Suborbital Starship

    It's clear that Starship can't reach orbit without a zero payload, nor would it be able to land safely back on Earth.  Single stage to orbit (SSTO) spaceships have been a recurring dream in the space industry since it began, and of course, they are a staple of science fiction.  So far, none has been built.  The physics is just too hard.

    SpaceX's original plan for point-to-point hyperfast travel on Earth was to use a giant booster, called Super Heavy, to launch the Starship upper stage into orbit. 

    But then, a year ago, Musk talked about a Starship doing suborbital flights, capable of reaching Mach 20 (20 times the speed of sound/24,700 kph/15,500 mph), travelling up to 10,000 kms, and carrying 450-500 passengers.  These suborbital flights would not need the Super Heavy booster, which makes the whole idea far more practical. On the other hand, a flight length of 10,000 km won't cover the journey from the USA to Australia, though it is enough for L.A. to Tokyo and L.A. to London.  Most cities on Earth could be reached with one or two flights.  This will be potent competition for long range air travel.  Passenger jets fly at speeds below Mach 1.

    That brings us to five slightly different versions of Starship: three LEO types (passenger, freight, fuel), a Moon version and the suborbital shuttle.  They'd all be variants of the same basic model.

    By the way, does the explosion of SN4 on the test stand mean that the dream of Starship is finished?  Definitely not.  Each "unscheduled rapid disassembly" has happened further along in the testing chain.  In this case, it wasn't the vessel itself which failed but a methane fuel line connection to the Starship.  SpaceX isn't just advancing down a construction learning curve, it's also going along a procedures learning curve.  You can be sure that they won't make this same mistake again.  Prototypes of Starship are being built so rapidly that there is already one completed vessel (SN5) and another half completed (SN6).  My expectation is that we'll still see 100 k "hops" this year [Update 10/7/21 -- orbital flights will take place later this year.  Elon time!]





    Sunday, May 10, 2020

    Starship full stack

    Image from Gravitation Innovation via HumanMars.  SpaceX is prob'ly going to tweak the design again (see the Moon version of Starship), and Musk is muttering about a wider span for the landing legs, but this is what it looks like for now.





    Wednesday, March 11, 2020

    Finally: an uneventful Starship test!

    Pressure test SN2
    Source: Boca Chica Gal/NASA Space Flight via Teslarati


    We all know Musk's methods by now:  test beyond breaking point, then improve, and test again, iteratively, until you get something that works.  Well, the latest Starship pressure test was 100% successful.  In fact, even better, the pressure test was conducted while the methalox tanks were being pushed up by a hydraulic jack to simulate the pressure caused by firing one raptor engine, thus showing it can handle the double stresses of rocket launch and overpressurised tanks.

    Teslarati comments:

    According to Elon Musk, SpaceX has successfully completed its latest Starship prototype test in a uniquely uneventful fashion, great news for the next-generation rocket’s next steps and first flight tests.

    The SpaceX CEO revealed the news some 12 hours after the company wrapped up the Starship tank test at its Boca Chica, Texas facilities. Another excellent example of SpaceX’s preferred process of agile development, the test followed just nine days after the Starship SN01 prototype’s first cryogenic test unexpectedly unearthed a design flaw. SpaceX analyzed the results of Starship SN01’s unintentional launch debut and drew up plans to rapidly repurpose a Starship tank initially destined for the SN02 prototype.

    By using existing hardware to test an upgraded iteration of the part that destroyed Starship SN01, SpaceX has now effectively retired the risk posed by that prior failure less than two weeks after it occurred. Elon Musk specifically noted that the former SN02 engine section “passed cryo pressure & engine thrust loads,” confirming that there was more to the exceptionally uneventful evening of March 8th than met the eye. While putting on much less of a show for local observers, this particular boring test is a great sign for the next few steps of SpaceX’s Starship development program.

    Musk’s description of the test suggests that SpaceX’s intention with the SN02 test tank – built in just two weeks – was to stress it up to (and likely beyond) the pressures and mechanical stresses Starship engine sections will need to survive in flight. In simpler terms, they likely tried to burst the tank by pressurizing it with liquid nitrogen, a supercool cryogenic fluid. It’s unclear exactly how far SpaceX pushed the tank, but it’s safe to say that it went at least as high as past test tanks, meaning 7-8.5 bar or 100-125 psi. At a bare minimum, a test that failed to reach Starship’s minimum flight pressure of 6 bar (90 psi) would be of dubious value for the actual orbital ship.

    A step further, SpaceX installed a hydraulic jack underneath the test tank in a bid to simulate the stresses it would experience with a single Raptor engine. Capable of producing approximately 150-200 tons (1500-2000 kN) of thrust, even Raptor is relatively minor compared to the Starship tank’s likely ~500 metric ton (1.1 million lb) mass. Still, the fact that the SN02 test tank survived the combination of a highly pressurized tank and the simulated thrust of a Raptor engine suggests that SpaceX is now ready for a more successful repeat of Starship SN01 testing.

    [Read more here]

    Just a reminder: SN2's tank and hull structure was constructed in just 9 days.  SpaceX is obviously rapidly building up to its interim target of one Starship per week.  Responding to a question on Twitter, "What's the path forward now? Static fire with SN3 and hop with SN4?," Musk replied, "Static fire & short flights with SN3, longer flights with SN4, but spooling up the whole Starship/Raptor production line is really what matters."  A production line which aims to produce 50 Starships a year at a cost of just $5 million each.  The low cost is only possible through using an assembly line.

    What will all those Starships be used for?   Originally, when Starship (then the BFS) was first proposed, Musk said that 6 ships would be needed for the first expedition to Mars.  But if the new stainless steel Starship costs just $5 million, about one tenth of the cost of a carbon-fibre composite BFS, 60 ships could be sent for the same cost (there's additional fuel cost, but it is small in the context of the capital costs.)  Even if these Starships are just used once, to fly to Mars, the total capital cost would be just $250 million.  Of course, there will also be the Super Heavy first stages, but they will be fully re-usable, so fewer of them will be needed.  (NASA's current plans to get to Mars estimate cost at $100 BILLION.  For 5 people.)

    In addition, there is the point-to-point market for suborbital travel on Earth.  Most places on Earth will be about an hour apart by suborbital SpaceX Starship shuttle.  Since spaceports won't be built close to cities, because of the noise and fears about rocket ship safety, the city-centre to city-centre trips will take, say, three or four hours.  But it currently takes 14 hours to fly from New York to Beijing, 22 hours to fly to Sydney (the Australian one, not the Canadian!)   There's surely a huge market for intercontinental shuttle flights.

    But wait!  Those are not all the potential markets.  Cheap launch to LEO (low Earth orbit) will mean for more space activity.  It used to cost $22,000 to lift  a single kilogram into LEO.  With the Starship/Super Heavy combo, that cost will fall to $20/kg, or lower.  Space tourism will become a thing.   And the cost won't be prohibitive.  Lifting a 100 kg person with 100 kg of food/luggage into orbit will cost just $4000.  Even with a 100% profit margin, that's still withing the range of anyone who now pays for intercontinental business class flights.  There will be millions who would like to take a "spacecation".  There will be rotating space stations for longer stays, financially feasible because of the low cost of lifting material into orbit.  And such low launch costs mean that a Moon base, or many Moon bases, will happen too.  Oh, and don't forget Starlink, SpaceX's global high speed internet service.

    Just as after WW2, when air travel plunged in cost and rose in quality, leading to an explosion of passenger air traffic, so will space travel when Starship is operating.  Starship will make its own market.  Every one of the Starships SpaceX produces will be needed.   In fact, there prolly won't be enough of them!

    Thursday, February 27, 2020

    Starship SN1 moves to test site

    (Images from SPadre)

    SN1 is moved to the test site.
    Note the dimple on the hull.  


    SN1 is lowered onto the test stand




    SN1 lowered onto the test stand


    SN1 on the test stand

    Note the dimple in the hull in the first picture.  Aircraft skin is 2-4 mm (1 inch = 25 mm) thick, but aircraft have internal bracing.  Is 2-4 mm steel strong enough for Starship?  Will Starship need internal bracing (struts)? 

    I'm certain it will.  The stresses caused by the "skydiver" re-entry manoeuvre will be more severe then those faced by aircraft. 

    But SpaceX is moving so fast, there'll be time to adjust the design.  It took SpaceX 9 months to build the first Starship prototype.  It has taken it just 4 weeks to build the next one, SN1 (serial number 1).  SN2 is already under construction.  The rapidity of progress on Starship compared with the huge delays and cost overruns of SLS ......  Astounding.  First boots on Mars in 2025!


    Wednesday, February 26, 2020

    Starship assembly line starts

    If Musk's claims that Starship will cost just $5 million to build are going to be right, then Starship will have to be built on what is effectively an assembly line.  He's planning to build 2 Starships a week.  And recent images from Teslarati show a nascent assembly line in Boca Chica.

    Nose cone construction at Boca Chica


    Tank bulkhead construction at Boca Chica



    Elon Musk has posted a new glimpse inside SpaceX’s South Texas Starship factory, revealing a nearly-completed rocket nosecone and indicating that the first upgraded Starship prototype’s flight debut is imminent.

    SpaceX teams have been working around the clock for a little over a month to build the first full-scale, flightworthy Starship prototype, a process that only began after two ‘test tanks’ were fabricated, assembled, and pressurized until they burst on January 10th and 28th. Built with improved tools and methods, those test results – particularly from the second test tank – allowed SpaceX to empirically confirm that its current infrastructure and techniques are ready to manufacture orbital-class (and even human-rated) Starships right now.

    And so work on the first truly flightworthy Starship prototype – known as SN01 (serial number 01) – thus began in earnest around mid-January, perhaps less than a month ago. Over the course of that month, SpaceX’s South Texas team has made spectacular progress. Starship SN01’s business half – comprised of a Raptor engine section, a liquid oxygen tank, a methane tank, and all associated tank domes and plumbing – is likely just a single big stacking and welding event away from being structurally complete.

    Without exaggerating, it’s safe to say that SpaceX has effectively gone from a handful of parts worth of Texas rocket production to a multi-vehicle, Starship production line concurrently manufacturing multiple vehicles in about eight weeks. While it would be theoretically easy for critics and a more general audience to see little more than some cheap stainless steel parts in a few hastily-constructed temporary tents, the reality is that SpaceX has already proven – at a minimum – that a steel Starship built with the exact same tools, facilities, and methods will likely be capable of spaceflight.

    SpaceX’s January 2020 Starship test tank program proved as much, demonstrating that thin steel tanks built in tents can serve as orbital-class pressure vessels and survive at internal pressures greater as high as 8.5 bar (125 psi) while filled with cryogenic (extremely cold) liquid. Meanwhile, Tesla’s Fremont factory General Assembly line 4 (GA4) – having continuously churned out high-quality Model 3s for more than a year – has proven that sprung structures can make for fast, cheap, and more or less permanent factory solutions. Prospective SpaceX competitor Blue Origin even based its own brand new headquarters – opened in January 2020 – around an odd U-shaped sprung structure.

    So the first flight of SN1 is perhaps weeks away.  Initially, SpceX will do 150 m "hops", then it will increase to 1 k "hops", steadily increasing to 20 k flights, where they will test the "skydiver" landing technique on the way back (and we will see whether SN1 can withstand the associated stresses).

    From a piece I wrote earlier (slightly amended):

    The SpaceX/Mars timetable still looks doable, unless there is a disaster:



  • H1 2020: Tests to 20+ kms altitude.
  • End 2020: Full stack (i.e., Super Heavy booster plus Starship) operational.  First orbital flights of the Starship.
  • Early 2021: First commercial customers (for satellites), launches of Starlink constellation
  • Late 2022: Uncrewed mission to Mars (Mars is in opposition in December)
  • 2023??First commercial space station.  Launched on Starship, built by non-SpaceX companies—or maybe even by SpaceX
  • 2023: 'Dear Moon' circumlunar expedition
  • 2024??: Moon Base Alpha
  • Early 2025: Crewed mission to Mars  (How many ships?  Here's my analysis)
  • Early 2027: Second expedition to Mars.  Return of at least one Starship.  
  • 2029: Third expedition to Mars.  Martian population reaches 300.  If Starship works, SpaceX will start designing even bigger rockets, capable of carrying more than 100 passengers, so Mars's population in 2029 could be more than 300.  Return of some Starships.
  • Sunday, February 23, 2020

    Zubrin talks Starship with Musk

    From an interesting YouTube video from What about it?, covering a conversation between Dr Robert Zubrin, president and founder of the Mars society, and Elon Musk, founder of SpaceX, which took place at the recent SpaceX employment fair in Boca Chica.


    1.  There are currently 300 employees at SpaceX's Boca Chica operations.  This will be rapidly increased to 3000.
    2. The new (again) LA facility will be used to build raptor engines, while Boca Chica will be where the rest of Starship and Super Heavy are assembled.  Musk plans to build 2 Starships a week.
    3. Each Starship will cost—wait for it—just $5 million.  (The Super Heavy will be extra.) My earlier guesses were 10-15 times as high.  This is extraordinarily cheap.  It confirms that SpaceX will be running an  assembly line for Starship/Super Heavy.  Musk's earlier estimate of $2 million per launch (i.e., both Starship and Super Heavy) seems very plausible.  9 launches will be enough  to get a Starship to orbit and to refuel it for the trip to Mars, which means (at 100 passengers) a Mars ticket cost of just $180K.  Just a reminder: SLS will cost $1.5 to $2.5 BILLION per launch.  For the same cost, NASA could send 8000 astronauts to Mars.  Just saying.  Meanwhile, the whole fleet of 1000 Starships would cost just $5 billion.  You'll need fewer Super Heavys because they'll be re-usable several times a day, whereas Starships will have a more limited re-use because they'll be en route to Mars, or on the red planet itself, but on the other hand, the Super Heavys will cost at least twice the cost of the Starship itself.  So $10-$15 billion for the whole fleet?
    4. The first 5 Starships will stay on Mars, which makes sense as they will be so cheap.  These will be the cargo versions of Starship, but there is no doubt they could be used as temporary habitats while permanent habitats are built.
    5. No nuclear reactors (he means, presumably, NASA's KRUSTY kilopower reactors).  Zubrin pointed out that 6 to 10 football fields of panels would be needed to refuel a single Starship, and Musk replied "Fine!  That's what we'll do."   I still think that will require more Starships than SpaceX has been talking about so far, just to carry the solar panels.  But if Starships are so cheap ....  On the other hand, the next settlers may well be happy to have a diversified mix of nuclear, wind and solar, given the reality of planet-wide dust storms lasting months at a time.
    6. The first crew ships will carry just 20-50 people, instead of the 100 each ship is capable of carrying.  Most prob'ly because  of the need for cargo space.
    7. Musk also dismissed the whole "mini Starship" idea that Zubrin has been promoting.  Starships will be so cheap and production so rapid that leaving a Starship on Mars for 2 years (the Mars-Earth orbits only "sync" every 26 months) will only add a small amount to total costs.   
    8. SpaceX will go for a 100 km high launch immediately after the 20 km launch.  Though he didn't say whether this would be SN1 (Serial number 1) or SN2 (both under construction at Boca Chica right now) which will attempt these landmark milestones.  100 kms (suborbital) will allow Starship to do point-to-point flights, and to circumnavigate the globe.  
    9. Zubrin thinks that it is very likely that SpaceX will put boots on Mars before NASA gets to the Moon.  Looks very likely, doesn't it?
    10. Progress constructing the first two test models of Starship, SN1 and SN2, has been extraordinarily rapid.  Subject to bureaucracy, the first test flights could happen as soon as April.


    Wednesday, January 8, 2020

    First F9 test flight to 1000m



    Video of Falcon 9 Reusable (F9R) taking its second test flight at our rocket development facility. F9R quadrupled its height from its previous test to rise to 1,000m.

    Early flights of F9R will take off with legs fixed in the down position. However, we will soon be transitioning to liftoff with legs stowed against the side of the rocket and then extending them just before landing.

    The F9R testing program is the next step towards reusability following completion of the Grasshopper program last year. Future testing, including that in New Mexico, will be conducted using the first stage of a F9R as shown here, which is essentially a Falcon 9 v1.1 first stage with legs. F9R test flights in New Mexico will allow us to test at higher altitudes than we are permitted for at our test site in Texas, to do more with unpowered guidance and to prove out landing cases that are more-flight like.

    This was just six years ago.  This week, SpaceX launched a Falcon 9 booster for the 4th time, and landed it safely.  This quarter, SpaceX will likely start flights of Starship, at first to 100 metres, then 1,000 then 20,000.  And if they all work as planned, we'll see the first suborbital flights later this year, and the first commercial flights next year.  Even if they don't work as planned, SpaceX will learn from their mistakes, and will rapidly improve Starship.  


    Tuesday, January 7, 2020

    SpaceX Starship orbiting Earth

    A great render from Gravitation Innovation.  Hat tip to Human Mars.

    We're likely to see the first low altitude (20 kms) tests by March or April, with orbital tests by the end of the year.  Crunch time.  If Starship works, it changes everything about space and our view of ourselves in the world.


    Wednesday, November 13, 2019

    Cheap access to space

    Starship Mk1 at Boca Chica 



    From Inverse:

    Starship, SpaceX’s ambitious rocket under development, could have an underestimated effect on society’s interactions with space.

    The stainless steel rocket is currently in the prototyping phase, but when complete it’s set to unlock some of SpaceX’s most ambitious targets. It offers a fully-reusable design, which could drop launch prices lower than ever to $20,000 per ton. It uses easier-to-harvest rocket fuel, which could create a planet-hopping network to the moon, Mars and beyond. It can also carry over 100 tons or 100 people into orbit at once, meaning humans can send more into space than ever before.

    All this adds up to a rocket that could smash current annual launch metrics, CEO Elon Musk explained via Twitter Thursday. Humanity currently sends around 500 tons into space per year. Around half of this is completed by SpaceX’s Falcon family of rockets, which covers the Falcon 9 and Falcon Heavy. But the Starship is designed to fly three times per day, or 1,000 times per year.

    Assuming SpaceX builds around 100 Starships, matching the number of Falcon rockets, that amounts to a potential capacity of 10 million tons of payload entering space per year. That’s 20,000 times more payload than the whole planet is sending to space annually right now.

    That’s a lot of stuff — and it could eventually pave the way for Musk’s big ambition to build a city on Mars.

    SpaceX has its own ideas for how it wants to use the added capacity. It plans to develop the Starlink, an internet connectivity constellation that could eventually number 42,000 satellites. Considering the total number of satellites in space is somewhere around 5,000 it could vastly increase the number of crafts in orbit. Using so many satellites at low altitude could enable faster response times and higher speeds, while enabling SpaceX to earn more money through offering internet connection services.

    The company isn’t limiting its satellite ambitions to its own endeavors, though. Starship’s first mission is expected to send a telecommunications satellite into space as early as 2021. Jonathan Hofeller, SpaceX vice president of commercial sales, said in July 2019 that the company will encourage others to get creative with Starship’s capabilities: “You could potentially recapture a satellite and bring it down if you wanted to. It’s very similar to the [space] shuttle bay in that regard. So we have this tool, and we are challenging the industry: what would you do with it?”

    Perhaps SpaceX’s most ambitious goal is to build a city on Mars. This could be enabled by a fleet of Starships, all lifting cargo to the planet.

    Musk explained that around 1,000 Starships would be necessary to start a city. The fleet would transport around one million tons of cargo to the planet, which Musk estimates would be enough to develop a self-sustaining city. That would take around 20 years, as the Earth and Mars align around once every two years.

    SpaceX has big plans to meet these lofty goals. It’s aiming to build one new Raptor engine every day next year, with each Starship vessel and Super Heavy booster using around 40 engines total. It’s aiming to build new Starship prototypes at both its Texas and Florida facilities.

    All of this is in service of a ship that could transport 100 humans into space at a time. Space industry venture capitalist Rick Tumlinson previously told Inverse that this could be a major step in humanity’s history.

    “Even if he doesn’t go to Mars or the moon, the ability to deliver 100-plus people at a time to orbit, to space itself, is going to be the beginning of the biggest revolution in the history of humanity, if not life itself,” Tumlinson said in February 2019. “Having the ability to climb out of the gravity well of Earth means we have the potential to make life and humanity immortal, unkillable.”

    In fact, in a presentation to the Mars Society this month, Paul Wooster of SpaceX said that the payload will be 150 tonnes, not 100.  Before SpaceX, it cost $10,000 to lift one pound into LEO (low Earth orbit).  That's $22,000 per kilo.  With Starship, that cost will fall to $13 per kilo. That is a cut of nearly 3 orders of magnitude (10,000 times).  Even if we never get to Mars—and we will, within a decade—this will open up space.  As Tony Seba says, every time there is a 10-fold fall in costs there is a disruption.  How big will the disruption be with a 10,000-fold fall in costs?  At 100 passengers, it would theoretically cost $2000 to lift a person into LEO.  How many people will want to see the world from orbit?  How many would like a "spacecation"?  Because it will cost only $13,000 to lift a tonne into LEO, there will be lots of commercial space stations.  Hoteliers will charge $5000 for a week in space.  There will be a booming business of holidays in space.

    What about the journey to Mars?  Well, the first expeditions will be expensive, because the Starships will be used as habitats on Mars, which means they won't be re-usable.  But at least some of the Starships that go to Mars for the 2027 and subsequent expeditions will be able to leave their cargo and passengers on Mars and return to Earth within the Mars-Earth opposition window, where they will be fully re-usable for ferrying people and cargo up to space stations in LEO, or to the moon, or for point-to-point travel on Earth.  The trip to and from Mars will take 7 or 8 months, including the couple of weeks on Mars for loading and unloading.  So for 2/3rds of the time, those Starships will be re-usable, back here on Earth.  Now, to send a Starship to Mars will require 8 refuelling launches from Earth, each costing $2 mill.  Add in the cost of fuel to get to Mars, and total cost is, say, a conservative $30 million, including the cost of the slightly-less re-usable Starship, which will obviously be higher than a fully re-usable one, though nowhere near as high as one which is used just once to fly to Mars.

    Remember that the Super Heavy booster will never go to Mars, and will be 100% re-usable, 3 times a day according to Musk.  Only the Starship part of the Super Heavy/Starship combo on the Mars trip will be partially re-usable and then for just a one third of the time.  At a conservative $30 million, the cost per tonne to Mars will be $200,000.  The cost per person weighing 100 kilos, with 100 days of food (we eat one tonne of food a year), making a total of (let's be conservative and assume 180 days of food) 150 kilos, will be $30,000.  Earlier, I estimated the cost of a trip to Mars in the carbon-fibre composite BFS at an optimistic  $100,000, and added:

    remember that it cost £30/US$150 to cross the Atlantic one way by steamer in 1900.  The average weekly wage in 1905 in the US was $10.05, so a one way ticket across the Atlantic cost about 1/3rd of year's wages. In 2014 the average wage in the US was $73,298. 

    Assuming a 2% nominal per annum increase in the average wage in the US, the average wage in 2027 will be just under $100,000.   So, with the new stainless steel Starship/Super Heavy combo, a trip to Mars will cost about 1/3rd of  the average annual wage in the US in 2027, just as it cost one third of the average wage for a trip by steamer across the Atlantic a hundred and twenty years ago.  How many immigrants poured into the US in the early years of the 20th C?  In the decade 1900 to 1909, 8.2 million people moved to the US, and in the next decade, despite a world war, 6.3 million.  Does anyone doubt that there will be hundreds of thousands of people willing to emigrate to Mars when costs for a ticket fall to below $30,000 or below?  I say $30K or below, because by 2027, there will be a new, improved version of the Starship which will be cheaper than today's version. Musk is already talking about an 18 metre wide version 2 of the Starship.   This will likely halve launch costs.

    At first, it will be just a few people, because the Martian economy will take time to get going.  We'll need jobs on Mars to soak up all the immigrants.  But as Musk pointed out, there will be a shortage of labour on Mars, meaning wages will, relative to Earth, be higher.  So of course, will living costs—Martians will have to pay for life support, because they will not be part of a self-sustaining biosphere.  Immigrants add to demand, creating the need for more labour.  So a feedback process will start, with numbers of Mars rising, requiring more labour, which means more immigrants, which means even more people on Mars which means more demand .......  There will be tourists, visitors to Mars. What would millionaires pay to fly over Valles Marineris, the deepest and longest canyon in the solar system?  To be able to boast back home about their visit to Olympus Mons, the tallest and largest known volcano in the solar system?  To walk on the surface of another planet?

    Just as the need for high-density, cheap, rapidly rechargeable batteries is now driving a technological explosion in batteries, so cheap access to space will drive a welter of new technologies, which will be essential in space but also very useful on Earth.  Techniques for extracting CO₂ from the atmosphere, cheaply and efficiently; ways to produce synthetic methane cheaply; food production (vat meat and hydroponics, for example); life support systems; things made in space (fibre-optic cables which contain no flaws because they're made in zero gravity); water purification systems; a huge increase in understanding the human body and its health; genetic modification and gene repair; communications; science .....

    We are going through several major technological revolutions—3D printing, access to space, the triumph of renewables, AI (Tesla self-driving), the EV revolution, communications (think Starlink, bringing high-speed broadband to the whole world).  All of these will change the world in ways none of us expect, but be sure: just as the internet and smart phones have revolutionised our society, these new revolutions will alter it beyond recognition.

    Thursday, November 7, 2019

    Cost of the Mars expedition

    Starship Mk1 at Boca Chica
    Source: TechCrunch, photo by Darrell Etherington


    Recently, Elon Musk has said that a single launch of the Starship/Super Heavy combo will cost just $2 million. 

    SpaceX’s  goal has long been to achieve truly reusable rocket launch capabilities, and for good reason: The company anticipates huge cost savings through re-usable rocketry versus expendable launch vehicles, which SpaceX CEO Elon Musk  has described as a process akin to an airline throwing away their passenger aircraft every time they complete a flight. They’ve made lots of progress toward that goal, and now frequently re-fly parts of their Falcon 9 rockets and their Dragon cargo capsules — but the Starship spaceship they’re building now should be even more re-usable.

    Musk provided an idea of just how much that could save SpaceX — and by extension, its customers — at a surprise guest appearance at the U.S. Air Force’s annual pitch day in LA this week. Speaking with USAF Lieutenant General John Thompson at the event (via Space.com), Musk said that fuel costs for the Starship should be around $900,000 per launch, and that once you factor in operational costs, it’ll probably add up to around $2 million per use. 
    [From TechCrunch]

    It seems likely that the early Starships will not return from Mars, but will instead be used as temporary habitats while permanent living modules are built on and below the surface of Mars to shelter the first settlers, provide places to grow food, create air, recycle water, and all the other things humans will need on Mars to survive.  So each Starship from the first few expeditions that go to Mars will not be re-usable, or not often enough to cut its cost, though the Super Heavy that launches it will be.  Musk (I think) also hinted that the capital cost of a Starship is less than the cost of an expendable Dragon 9, which is $62 million.  (In a previous piece, I suggested that the cost of each Starship may be as low as $10 million, which is prolly too low.  The stainless steel frame might be just 2% of the cost of a carbon fibre composite rocket, but the engines, life support, fittings, etc, will push this cost up substantially.  But it might nevertheless be way below $60 million.)

    So the cost of each Starship going to Mars will be, say,  $60 million for the Starship, plus $2 mill for the ten fuel launches to refill the Starship's fuel tanks in orbit, plus the cost of the fuel for the journey to Mars ($1 mill after rounding), plus the cost of the initial launch.  Less than $100 million.  SpaceX plans six ships to Mars of which four will be cargo and 2 crewed.  $600 million in total.  A single launch of the SLS will be between $1.5 and $2.5 billion per launch.  For an expendable vehicle.  And that won't give us any habitats on Mars, included with the SpaceX option.

    $600 million to put boots on Mars.  OK, there will be other costs.  Rovers, power sources (wind turbines/solar panels/NASA's nuclear generators), food for at least 2 years, and of course, there will be a third expedition in 2027.    And although we would all like the first crewed ships to carry 100 people as the first ships did in Red Mars (it seems just right somehow, and, spookily, the arrival of the first Starships on Mars will happen 30 years after the first publication of that seminal work, published in 1992), the number will prolly be much lower, the minimum being set by the needed skillset for starting human life in a new and very hostile environment—engineers, biologists, gardeners, technicians, doctors, electricians, scientists, builders, robot techs, and more I don't know about.  Maybe, after all, we'll end up with the "first hundred", split between two ships.  Or maybe we'll have more ships, because we'll need so much food until the greenhouses and protein vats get going.  Each person eats roughly 1 tonne of food a year, so for 100 people, we'd need 200 tonnes of food, perhaps 300 just to be on the safe side.  That's two Starships for food alone!  Add in the cargoes of power generators, and you get to 4 or 5 Starships just for food and power.  So in all probability, the 2025 expedition will need 7 or 8 ships for cargo alone, as well as the crewed ships.

    Despite all the extras costs, we're talking about not much more than $2 billion total for the 2022, 2025 and 2027 expeditions, which works out at $250 million  for each year between now and 2027.   That's just over 1% of NASA's annual budget of  $21.5 billion.   If NASA won't pay, the Air Force might.  And if they don't, I suspect SpaceX will have enough resources to do it itself, if no one else wants to play.  In fact, Musk could pay for it personally.  His net wealth is north of $20 billion.

    SpaceX might still fail.  Safely re-entering a vessel as big as the Starship into the Earth's atmosphere will be an enormous technological and engineering achievement.  But remember when rival space companies fell about laughing when SpaceX tried its first retro-propulsive landing and failed?  Now, just 3 years later, it is routine.  Something that no national space agency or private company had ever achieved.  And SpaceX achieved it.  Somehow, I am very confident that in 2022, the first uncrewed Starships will land on Mars, followed by the first astronauts 26 months later.  I have to live just another 6 years to see humans on Mars.  I might just do it.






    Tuesday, October 29, 2019

    Starship's game-changing launch capabilities

    Starship's payload bay opening to release Stralink satellites.
    Source: Teslarati



    From Teslarati:

    SpaceX President and COO Gwynne Shotwell teased new information detailing the wealth of benefits that the next-generation Starship launch vehicle could bring for the deployment of the company’s Starlink internet satellite constellation.

    Beyond Shotwell’s clear confidence that Starlink’s satellite technology is far beyond OneWeb and years ahead of Amazon’s Project Kuiper clone, she also touched on yet another strength: SpaceX’s very own vertically-integrated launch systems. OneWeb plans to launch the vast majority of its Phase 1 constellation on Arianespace’s commercial Soyuz rockets, with the launch contract alone expected to cost more than $1B for ~700 satellites [1.4 million each].

    SpaceX, on the other hand, owns, builds, and operates its own rocket factory and high-performance orbital launch vehicles and is the only company on Earth to have successfully fielded reusable rockets. In short, although Starlink’s voracious need for launch capacity will undoubtedly require some major direct investments, a large portion of SpaceX’s Starlink launch costs can be perceived as little more than the cost of propellant, work-hours, and recovery fleet operations. Boosters (and hopefully fairings) can be reused ad nauseum** and so long as SpaceX sticks to its promise to put customer missions first, the practical opportunity cost of each Starlink launch should be close to zero.

    In a perfect scenario, the only material cost of Starlink launches should be the satellites themselves and each expendable Falcon upper stage, which SpaceX has no plans to recover. Speaking prior to Starlink’s 60-satellite “v0.9” launch debut, SpaceX CEO Elon Musk stated that each prototype spacecraft ended up costing more to launch than to build, despite the fact that their first launch flew on a twice-flown Falcon 9 booster.

    Musk thus implied that each Starlink satellite likely already costs significantly less than $500,000 even before SpaceX has begun to reap the full benefits of economies of scale. In fact, based on official 2016 figures that estimated the cost of each BFR booster/ship at less than $4M [it will be lower now, with a stainless steel spaceship] and Musk’s estimate that Starship could cut Starlink launch costs by a factor of 5, the cost of Starlink v0.9 production could have actually been as low as ~$350,000 apiece, with launch costs on the order of ~$20M.

    Speaking a little over five months after Musk, Shotwell revealed that a single Starship-Super Heavy launch should be able to place at least 400 Starlink satellites in orbit – a combined payload mass of ~120 metric tons (265,000 lb). Even if the cost of a Starship launch remained identical to Starlink v0.9’s flight-proven Falcon 9, packing almost seven times as many Starlink satellites would singlehandedly cut the relative cost of launch per satellite by more than the 5X figure Musk noted.

    In light of this new figure of 400 satellites per individual Starship launch, it’s far easier to understand why SpaceX took the otherwise ludicrous step of reserving space for tens of thousands more Starlink satellites. Even if SpaceX arrives at a worst-case-scenario and is only able to launch Starship-Super Heavy once every 4-8 weeks for the first several years, that could translate to 2400-4800 Starlink satellites placed in orbit every year. Given that 120 tons to LEO is well within Starship’s theoretical capabilities without orbital refueling, it’s entirely possible that Starship could surpass Falcon 9’s Starlink mass-to-orbit almost immediately after it completes its first orbital launch and recovery: a single Starship launch would be equivalent to almost 7 Falcon 9 missions.

    [Read more here]

    I talked about the synergy between Starlink and the BFS/Starship here and here.  It's ironic (and wonderful) that private enterprise will get us to Mars long before governments will, and with minimal government funding*, because the company has invented a new way to bring high-speed internet to every corner of the globe, which will fund the Mars expeditions and will thereby open up the solar system to trade and settlement, but which in turn depends on cheap launches.

    The 100-fold-plus decline in costs brought about by SpaceX will also make space stations in Earth, Mars and Lunar orbit possible and practical.   I have no doubt that in 20 years' time, there will be several large privately-owned space stations in orbit round the Earth, at least one in orbit round Mars, and two or three thriving settlements on Mars.  The official timetables don't even have the Mars expedition until 2040.  It'll happen much faster than that, if my projected Mars timetable is kept to.


    *I expect NASA, ESA, RosCosmos, and other state space agencies will be happy to buy berths on SpaceX expeditions to Mars at the same rates that private individuals will pay, i.e., something like $240K (or maybe less). On the other hand, NASA's estimate when Congress asked how much a Mars expedition would cost was $15 billion for 5 astronauts.  And we all know the likely final cost would have been double or triple that.  Just look at SLS.

    ** Actually, ad nauseam—learning Latin at school has some use, after all.

    Sunday, September 29, 2019

    Musk's 2019 Starship presentation

    At Boca Chica Texas, right next to the silvery finished (almost—that'll take another month) Starship Mark 1, lit by searchlights, Musk shared his ideas about the future with us as well as providing us with new renderings of Starship.

    The new look stainless steel Starship.  Big fins at the back are (mostly) not for lift but to slow down the descent,
    like a skydiver.  The aim will be to slow re-entry enough to reduce heat, so some small lift in the upper atmosphere.
    Small nubs are landing leg housings.
    Steel costs 2% of carbon fibre composite, is much better at cryogenic temps and also at high temps.
    Is also re-usable in an emergency and can be re-welded on Mars/the Moon if needed.
    The reason they're being built outdoors is that putting up buildings would have taken too long.

    The three sea-level engines (with the smaller bows) are in the centre.  They can
    gimbel by up to 15%.  The outer three are for deep space travel, and don't gimbel. 
    Musk said in a tweet that there were 6 landing legs. 
    Presumably the two not shown are buried in the fins right next to the body.

    Starship entering Mars's atmosphere.
    Heat shield made of small ceramic tiles (material looking like spaghetti under the microscope???)

    Use of steel which has melting point of 1500 degrees C permits thinner and lighter
    heat shield.  Carbon fibre composite and aluminium melt at 400 degrees.

    Super heavy also of stainless steel.  Little fins at the base serve little aerodynamic purpose.
    They're housings for the landing legs.

    Height of Starship 50 m, of Super Heavy 68 m.

    Super Heavy returning to launch pad.
    Musk says aim is to re-use both Starship and Super Heavy within 6 hours of landing.

    Starship at Moon Base Alpha.
    Musk thinks settlement on the Moon will be unlikely,
    but permanent scientific bases as with Antarctica will exist.

    Not just our moon, but also the moons of Jupiter and Saturn.

    But colonisation of Mars remains primary goal.  We may be the only places with sentience in our galaxy.
    There's no evidence of alien civilisations (The Fermi paradox). And it took 4.5 billion years for sentience to emerge here.
    If it had taken just 10% longer, it would have been incinerated as the sun heated up over the next
    500 million years.  Implication:  we have duty to spread life and sentience through the universe.
     
    I'm doing this from memory, as the livestream video won't play, now that the presentation is over.  No doubt, SpaceX will upload the video in the next few days.  His timetable is as usual tight:

    • Starship Mk1 flying to 20 km (65 K feet) within a month or two
    • Starship Mk 3 and 4 built within 6 months.  These will be full sized.  Can't reach orbit by themselves, so Super Heavy must be built first.  Expects that to happen within 6 months, so first orbital tests next year.
    • Is only devoting less than 5% of SpaceX's resources to Starship.  However, cost of testing and iterative development much more expensive than building the prototypes.  Still focused on getting Crew Dragon working ASAP.  
    • Expects that they will build at least 10 Starships.  Total annual launch capacity (because they'll be rapidly re-usable) will be 50 times larger than total existing launch capacity, including SpaceX's existing fleet.  (I see capacity is back up to 150 tonnes from 100 in the previous version)  Unstated implication is that launches with Starship/Super Heavy will be super cheap, else where will demand come from? 
    • Will use both Boca Chica and Florida for launches.  Was cagey about what Boca Chica would look like in 10 years, but it's obvious: a massive spaceport.
    • Because both Super Heavy and Starship will be rapidly re-usable, testing will be able to proceed much more rapidly and much more rigorously.  With expendable rockets you'll need 10 rockets to test the design 10 times.  With Starship, you'll only need one.
    • Will test uncrewed first.  So first Mars and Moon missions will be automated and uncrewed.
    • The biggest difficulty he sees in the design and construction of the spaceships.  Life support, for example, he sees as a much easier.
    • Will ultimately make methane on Earth as well as Mars using the Sabatier method.  Needs 3.5 tonnes of oxygen for each tonne of methane and will extract the oxygen near the spaceport, rather than shipping the huge volumes needed in.
    • Rapid technological and manufacturing progress.  For example, will no longer cut the steel into squares, but will just unroll it from the rolled steel coil and shape it to the circumference of Starship with a single cut and weld.  Hence even more rapid timetable than we have seen to date. Will have to step up pace of Raptor engine manufacture from one every 8 to 10 days to one a day by or before Q2 next year.
    My conclusions?  SpaceX is very much on track to achieve its timetable.