back to article SpaceX's internet satellites to beam down 'Hello world' from orbit

The first two internet-relaying satellites in SpaceX's Starlink constellation have been launched into Earth's orbit – and will begin broadcasting to the world this week. A refurbished Falcon 9 rocket carrying the two birds, as well as the Spanish Paz radar satellite, blasted off today from Space Launch Complex 4 East at …

  1. Stevie


    They need to *fire* the net at the fairing and reel it in after both hit the drink. Four rockets where those posts are arranged to spin the net slowly as it hurtles at the skittish and uncooperative fairing.

    Or they need a fleet of recovery mini-subs in addition to the boat.

    Or something like Sky One that can fly up to the fairing and nab it in a net that the submarine part reels in and salvages.

    1. Jon 37

      Re: Bah!

      Salt water is very corrosive, which is bad for your only-just-strong-enough metal rocket parts. SpaceX have been really careful with most of their recovery efforts to ensure things don't get dunked in water. The exception is Dragon, which lands in the sea, and the only "reused" Dragon capsule had to be completely stripped down and rebuilt with mostly-new parts, which is very expensive.

      So their plan to catch the fairings *before* they hit the water probably leads to *much* cheaper refurb costs than fishing them out. But it will take a few more attempts to get it working, then a few more to get it mostly reliable. Any capture method will never be 100% reliable.

      1. Yet Another Anonymous coward Silver badge

        Re: Bah!

        The fairing are carbon fiber

        Of course it is still better to catch them in a net than try and fish them out of the water and haul them aboard

        1. lglethal Silver badge

          Re: Bah!

          I must say I would not want to be on board a ship thats trying to catch with a net a big old piece of heavy space equipment travelling at mach 8! What the hell do you make a net out of that can catch something travelling at mach 8!

          icon --> for what happens when you get the catch wrong! :P

          1. ArrZarr

            Re: Bah!

            If I might suggest that it'll only be travelling at mach 8 when it starts its reentry as opposed to screaming through the whole atmosphere at that speed. It'll probably be doing nice, leisurely motorway speeds by the time it's caught by the most gloriously weird ship I've seen in a while.

            1. lglethal Silver badge

              Re: Bah!


              I'm interested how you think they plan to slow it down. They've said no parachutes involved and it doesnt come with a heat shield so they cant use aggressive levels of aerobraking.Just using standard atmosphere to slow it down but without making it burn up, wont slow it down that much from Mach 8 (my very rough guess - its a friday and im off to the pub so im not going to actually work it out - would be Mach 3), certainly i dont think it would be down to "leisurely motorway speeds."

              On that topic though, I'm also interested in what sort of net you would like to deploy to catch the equivalent of a car travelling at motorway speeds? If you have something, please contact Top Gear - I guarantee they would be interested (and maybe the police for car chases, but Top Gear would certainly be more entertaining! ;) )

        2. Brangdon

          Re: The fairing are carbon fiber

          The fairings are partly carbon fibre, and partly aluminium and other things.

      2. Stevie

        Re: Bah!

        Spreadsheet or shenanigans called. NASA fishes stuff out of the sea all the time, or did.

        None of the Mercury, Gemini or Apollo capsules that I have examined up close appears to show any salt water problems after spending time marinading. You can still see the scorch marks from re-entry on certain parts, but no salt corrosion to speak of. They only display the Liberty Bell 7 in semi-darkness, so it's difficult to tell what years resting on the sea bed did, but that of course is a completely different use case.

        And I'll bet that being soaked in hydrazine vapour, rained on, baked in the Florida sun then smashed about at escape velocity through the air is a lot rougher on the materials than a bit of sea water for a few minutes.

        But when you show me the figures I'll stand corrected in what was obviously intended to be a light-hearted suggestion in the first place.

  2. jake Silver badge

    One wonders ...

    ... what kind of latency we'll see with starlink.

    1. pdh


      Figure roughly 1000 miles each way for low-earth orbit, depending on the exact orbit height... speed of light is about 186,000 miles per second, so 2000 miles takes about 11 milliseconds -- that is, it would take that much longer for a one-way data packet from ground station 1 to satellite to ground station 2 as compared to a perfect zero-latency link. If the internal processing in the satellites can keep up without introducing more delay, that's not too bad.

      1. Adrian 4 Silver badge

        Re: Latency

        Wouldn't it have to hop through the constellation a bit, too ?

        1. Jon 37

          Re: Latency

          > Wouldn't it have to hop through the constellation a bit, too ?

          Depends on how far you are from a ground station. If speed-of-light travel time to your closest ISP is X milliseconds, then you're never going to get a latency less than X milliseconds by any means. The satellite latency will be the total of: around X milliseconds to hop around the constellation plus 11 milliseconds to go up & down plus a bit more for processing in each satellite and the ground station(s). Fiber would give you a latency of about 1.5 times X milliseconds, since light is slower in a fiber than through air/space.

      2. Anonymous Coward
        Anonymous Coward

        Re: Latency

        Possibly a lot less. Looking at the number of satellites in the lowest orbit you should never be more than 300 miles from one.

      3. Anonymous Coward Silver badge

        Re: Latency (@pdh)

        "Figure roughly 1000 miles each way for low-earth orbit"

        As a reference, the ISS orbits at approximately 200 miles altitude.

        With the numbers of satellites proposed there won't be much lateral distance, so even 500 miles would be a big overestimate.

      4. rh587 Silver badge

        Re: Latency

        LEO is anything less than 2000km (1200mi, 14.28million linguine), so your calcs on 1000mi are broadly "worst case".

        In principle it could be significantly less than 11ms, and if you're going direct house>datacenter without touching (other) ISPs, Tier 1s or IXPs, then you could shave off a little bit of intermediate routing time too.

      5. Jaybus

        Re: Latency

        In the end, the latency seen on the new service will depend more on Musk's ability to negotiate with the major network carriers than on satellite distance. The important bit is that this satellite Internet service should be much lower latency than existing satellites services using geosynchronous satellites that are more than 35,000 km away, so 230 ms instead of 11 ms. Quite a difference.

        The real culprit is going to be the ground-based switching. With more than 12,000 satellites, there will be 10s of millions of packets per second that need to be transferred onto the optical network(s). They are going to need fat pipes to the optical networks of existing carriers.

        Also, while the US FCC appears to be on board, they are going to need to deal with numerous such regulators globally. I should think that they can get all 12,000 birds up faster than they can negotiate with all the World's RF regulatory agencies, some of which will no doubt nix their choices of frequencies as a "negotiation" method.

    2. Anonymous Coward
      Anonymous Coward

      Re: One wonders ...

      If it's less than 394 ms, which is what Comcast is providing right now, it'll be a f---ing improvement. Satellite is the only other option here despite being with Fresno, Ca, city limits. AT&T doesn't even bother, not that I'd ever use them. I'd rather walk to the library and walking ain't fun anymore.

      1. TSG

        Re: One wonders ...

        I seem to recall someone having said the latency will ostensibly be around 110-200 ms. I can’t remember where I saw that though, so take my words with a grain or ten of salt.

        1. Anonymous Coward
          Anonymous Coward

          Re: One wonders ...

          The 110-200 msec figure is for geostationary or geosynchronous orbits. They're something like 25,000 miles away, the SpaceX birds are much closer.

          1. Anonymous Coward
            Anonymous Coward

            Re: One wonders ...

            Tell me how you have a scalable algorithm to find paths through a random network, knowing only adjacent nodes.

            Unless the source and destination are both with the footprint of the satellite, it will need to communicate with other satellites and find a path through it. And since the satellites are moving there will be a need for a lot of updates or some other form of management.

            This is probably solvable, but it hasn't yet been solved. So this could be interesting.

            1. Neil Barnes Silver badge

              Re: One wonders ...

              I suspect the fact that you'd have an ephemerides of the available satellites 'simplifies' matters - though I suspect you'd have to use routing software working on the principles of satnav routefinding rather than current gateway protocols.

              1. Anonymous Coward
                Anonymous Coward

                Re: One wonders ...

                @Neil Barnes this is a nice approach - it may not provide the optimal solution but one which is "good enough". This is probably what will happen.

            2. Jonathon Desmond

              Re: One wonders ...

              But it’s not a random network. Satellites move in fixed and predictable ways and clients generally won’t be moving fast enough that tracking them would be too challenging.

              At any given point in time, any given satellite will know Its own location, that of all the other sats in the constellation, and where the fixed ground stations (exit nodes) are. That would seem to me to make client to internet routing straightforward.

              I imagine that clients would be assigned some kind of identifier or address that indicates their physical location, or that information would be wrapped in the protocol somewhere, so that responses can be directed to the nearest satellite for the downlink. Slight variations and outages would be handled because, as with GPS, the client nodes would likely be listening to multiple overhead sats at once, so would receive the data intended for them regardless of precisely which nearby bird down linked it.

            3. Alan Brown Silver badge

              Re: One wonders ...

              "since the satellites are moving there will be a need for a lot of updates or some other form of management"

              There are already best-path routing algorithms in existence which converge in sub-millisecond periods. (eg OSPF and TRILL). This is a fairly solvable problem given enough computing horsepower in orbit.

          2. SloppyJesse

            Re: One wonders ...

            Good article on these constellations over at IET


            They claim latencies of around 30 ms versus 700 for geostationary satellites. Presumably that is just signal transit time.

            Interestingly they suggest spacex will start launching in 2019 and the article was only published last month. Wonder if they've stepped up the pace to try and catch up with other proposed networks?

    3. Alan Brown Silver badge

      Re: One wonders ...

      It's an interesting question.

      Back in the bad old days of having to use geostationary birds for international internet, "awful" was a pretty good summary.

      This should be much less (2-500 miles vs 34,000 miles) BUT there are a lot more hops to traverse and the second can pretty quickly overwhelm the first.

      This is the kind of solution where a low level constellation uses a midlevel constellation that actually talks to the ground. The problem with relying on ground stations within sight of the low level birds is that in a lot of cases they're going to be over countries where the "awful connectivity" is by design rather than by topography.

      China in particular isn't going to take kindly to the Great Firewall being backdoored in this manner.

  3. Ian Johnston Silver badge

    Litter louts

    For this flight SpaceX didn’t try and recover the Falcon first-stage booster and let it crash into the ocean. The rocket was one of SpaceX's older models and had already flown once, but the company is now only recovering the last generation of its rockets, the Falcon 9 "Full Thrust" v1.2.

    Since they could have recovered the stage if they wanted to, why the hell should SpaceX get away with dumping a huge lump of rubbish like that at sea? Antisocial bastards.

    1. Sorry that handle is already taken. Silver badge

      Re: Litter louts

      There are hundreds, probably thousands, of spent rocket bodies at the bottom of the ocean off the coasts of Florida and California.

      And you're having a go at the only company so far that has done anything to reduce this practice.

      1. This post has been deleted by its author

    2. Mark 85 Silver badge

      Re: Litter louts

      Not just rockets. but ships with cargo, dead fish, and even airplanes. Oh.. and humans, dead ones, but humans. SpaceX is at least trying and having success at recovering their rockets. Not just from a pollution standpoint but there's a savings in resources. etc. They should be commended not called "louts".

      1. Jellied Eel Silver badge

        Re: Litter louts- Worse things happen at sea

        Not just rockets. but ships with cargo..

        .. which were sometimes rockets. And bombs. And in one case a few years ago, a chap in the US ordered some gravel for his drive. He noticed a lump of metal, then noticed it looked like a shell case, which turned out to be a chemical munition. Lots got dumped at sea after WW2 and later, and can reappear today-

        1. Dave 126 Silver badge

          Re: Litter louts- Worse things happen at sea

          Sometimes old aircraft (with aluminium fatigue) and tanks are deliberately dumped in coastal waters to provide a structure for coral to grow on. I can't see an aluminium rocket stage in deeper waters being that bad for sea life.

    3. rh587 Silver badge

      Re: Litter louts

      Since they could have recovered the stage if they wanted to,

      They actually couldn't - the West Coast drone barge (JRTI) is currently laid up with most of it's station keeping equipment on the deck for maintenance where it's been for the last couple of months.

      SpaceX currently have no Recovery Capability for launches from the West Coast (i.e. Vandenberg).

      That's why they used an old Block 3 booster that they could expend instead of a shiny new Block 5 that they'd want to recover/refurb/reuse. The Block 3s are only good for 2-3 launches anyway so this core was EOL. If they'd recovered it, it would have been destined for a museum, not reuse.

  4. Sorry that handle is already taken. Silver badge


    Swiss manufacturer RUAG, which builds the fairings for Ariane 5 and Atlas V500 rockets, also has a plan to recover and reuse rocket fairings. Its plan is slightly simpler, just bundling in a parachute to the housing, and the first system is due to be tested later this year.

    SpaceX has had problems with this idea because, according to Musk, the parachutes don't work very well in the air stream coming off the fairing. Obviously RUAG's designs will be different and that often counts for a lot when aerodynamics is concerned, but I'm interested to see how they approach the problem.

    1. Martin Budden

      Re: Parachutes

      I assume you'd use a small drogue to slow it down, then maybe a larger drogue if needed, before deploying the main parachute.

    2. Alan Brown Silver badge

      Re: Parachutes

      I assumed when I read this that SpaceX was using a parafoil due to their steerability. (This is the same reason that Dyna-Soar plans in the late 1950s proposed using Rogallo airfoils and these were considered for the Mercury/Gemini/Apollo capsules before water landings were decided on.)

      One of the more interesting properties of a payload fairing is that it's a natural lifting body, which means that even if initially tumbling they're going to naturally align themselves for stable reentry as long as the centre of pressure and centre of mass are arranged appropriately. I wonder how much penalty there would be in adding control surfaces.

      IIRC the one that washed up on the Isles of Scilly showed no sign of burn damage, so they must be light enough that deceleration energy is very low.

      1. onefang

        Re: Parachutes

        "I assumed when I read this that SpaceX was using a parafoil due to their steerability."

        No need to make that assumption, the article said - "The new fairing has a parafoil added,"

  5. mt_head

    Scaled-back market

    >> the biz said it would be primarily focused on areas that don’t have good internet access, or where speeds are slow.

    So... Los Angeles, then.

    1. TSG

      Re: Scaled-back market

      I think you meant "post-net neutrality America"

  6. Tom Paine

    ...12,000-odd low-Earth orbit satellites ...

    That's either a typo, or the end of astronomy, or the mythical run-away orbital debris feedback loop.

    (Or 2 and 3: they're not mutually exclusive.)

    1. Francis Boyle

      Yes, 1200

      100 to 500 kg satelites according to the Wikipedia link (a good article - thanks Elon) all designed to deobit at EOL. And you thought the BFR was just for Mars.

  7. Martin Budden

    How long will they stay up?

    Low earth orbits tend to degrade because there's still a trace of atmosphere. That's a lot of satellites to replace every few years.

    1. I ain't Spartacus Gold badge

      Re: How long will they stay up?

      If they're small, and cheap, enough - SpaceX might chuck 10 or so on every single launch that's not up to max weight. If you're producing them in those stupid numbers, it probably works out cheaper to build more than it does to make the, last longer. At least, if you happen to own a rocket company...

      I think he said the fuel for a Falcon IX launch, is only $300,000. So once you've re-used a rocket a few times, and it's paid for itself and is perhaps reaching end of life, you might get a few more cheap uses out of it.

      1. Anonymous Coward
        Anonymous Coward

        Re: How long will they stay up?

        Take a deep breath, grab a cup of tea and sit down to go through numbers. Yes, it is boring, but sometimes very instructive.

        Now, think about the daily costs of operating the network and where is this money going to come from when the government stops subsidizing it.

        1. I ain't Spartacus Gold badge

          Re: How long will they stay up?

          Take a deep breath, grab a cup of tea and sit down to go through numbers. Yes, it is boring, but sometimes very instructive.

          OK. Where are these numbers? I'm presuming you've done them - as your post concludes that the thing isn't workable?

          Had a quick look at the wiki article on this, which appears to have had a lot of input by SpaceX. They're talking satellites of 100-500kg. Small enough to piggyback a couple on other launches. Even if we assume nearer the top 500kg weight, that means a Falcon 9 might lift 50-odd. Falcon Heavy being able to do 3 times that.

          BFR being able to lift 6 times that at 150 tonnes to LEO. And be re-usable of course.

          OK we have to also factor physical size into this. And how you deploy the damned things in multiples of 10. It also matters what size the constellation has to be to become viable. If you need 5,000 up before you can sell to the first customer, then you're in trouble. But if you can deploy say 500 and go operational then you can test the system and start getting revenue. So long as you don't sign up more customers than you have bandwidth. Say you could get 400 up in 5 launches that you pay for (on re-used rockets that are therefore essentially free), and another 100 piggybacking on the 20-30 launches a year SpaceX do for paying customers. Those 4 launches are going to cost you money, but we're talking a few millions to SpaceX - rather than half a billion to a customer - though there's the opportunity cost of what they could sell those rockets for...

          Now let's look at revenue. 10m US customers, paying $600 per year ($50 a month) = $6 billion per year. And as they're not in geosynchronous orbit, they get global coverage out of this network, so can make money elsewhere.

          Your infrastructure costs are incredibly cheap. Because you don't have vast numbers of engineers and a huge network and property to maintain. You just build loads of disposable-ish satellites - and they can be relatively cheap because you're mass producing them and they don't need to be engineered to last stupidly long.

          Is this viable? I've no idea. I'd have thought that 500 medium sized satellites would be better than 12,000 smaller ones. There's going to be global objections to launching a constellation that huge, covering that much of the sky. But I'm sure it's possible to make the numbers add up. And I'd imagine that SpaceX probably won't struggle to raise money for any reasonable idea they can come up with, given their past successes.

      2. John Brown (no body) Silver badge

        Re: How long will they stay up?

        "If they're small, and cheap, enough - SpaceX might chuck 10 or so on every single launch that's not up to max weight. "

        That's what I was thinking too. There's probably a little excess payload space on many launches and if you own the rockets, there's little extra cost in using up some of that excess capacity.

    2. Sorry that handle is already taken. Silver badge

      Re: How long will they stay up?

      I understand SpaceX aims to get its reusability to the point that the fuel is a significant proportion of the total launch cost, while building thousands of small (and possibly expendable) satellites will benefit from significant economies of scale.

      It's certainly ambitious, but with what they've already achieved I'm willing to give them the benefit of the doubt.

  8. Anonymous Coward
    Anonymous Coward

    Has Elon, peace upon him, done the economics of 12,000 LEO satellites? These things have a life of less than 5 years, which means that SpaceX will need to be launching more than 2400 satellites a year or about 6 to 7 per day, regardless of the weather.

    With grasp of manufacturing and logistics that Elon, peace upon him, has demonstrated to date, would SpaceX be able to manage this?

    Also, what impact will 12,000 LEOSATs have on other activities worth pursuing, such as Visual Astronomy? Or those of us to go on walkabout in the desert be far away from this sort of thing.

    1. Brangdon

      The initial constellation is only 4400 satellites. They have a lifetime of 5 years, so will probably deploy over the 5 years starting from a year or so's time, and then be replenished at 20%/year. Works out at about 17 a week. We don't know how many can be deployed in a single Falcon 9 launch, but it could be 20. Falcon Heavy could do more; they will be limited by volume rather than mass. I suspect the second constellation of 7,000 satellites will need the BFR to make it viable.

    2. Anonymous Coward
      Anonymous Coward

      > what impact will 12,000 LEOSATs have on other activities

      SpaceX will "own" space, nobody leaves the planet without permission, or they will find a small satellite blocking their way. ALL HAIL ELON MUSK, Master of the Universe!

  9. Anonymous Coward Silver badge

    Whining about Elon

    So many comments along the lines of: I bet he hasn't thought about x. What about y? Why doesn't he do z?

    Look folks, he's a multi-gazillionaire with some massively successful businesses along with a few 'fun' projects that he funds himself. He knows how to do business and how to make money (and how to spend it!)

    Yes, I'm sure he has thought about whatever triviality you're suggesting.

    He's also great at coming up with new concepts and new techniques for doing things.

    If he says he can do it, then he's already thought it through and knows that he can.

    Like that Australian battery pack, for example - ostensibly really tight timescales and a massive project that he flippantly agreed to and delivered.

    1. Alan Brown Silver badge

      Re: Whining about Elon

      "If he says he can do it, then he's already thought it through and knows that he can."


      Note the differences between when he's talking about stuff he hasn't nailed down and stuff that he has. He doesn't take bets unless he already knows he's got the kit on hand to win them (vs having done it in the lab and not in the real world).

    2. This post has been deleted by its author

    3. Anonymous Coward
      Anonymous Coward

      Re: Whining about Elon

      Because the South Australian battery pack was a stunt that was need by the premier Jay Wetherall to save his political career. It was rammed through with little or no real testing, and on the figures, it provides very little benefit.

      Some of us will be paying handsomely for this little stunt, which served only to stroke Elon's over-inflated ego and to save a political incompetent.

      As with almost everyone else who made a fortune in Silicon Valley and elsewhere, Elon got lucky - that is why he made it. As for his aptitude in running a serious business - how is Tesla doing?

  10. Anonymous Coward
    Anonymous Coward

    Tintin A & B

    Surely it should be Thomson and Thomson (or even Dupond et Dupond)

  11. Anonymous Coward
    Anonymous Coward


    I assume that SpaceX had to rename the system to StarLink (as MS had to with SkyDrive) due to dangers of a trademark dispute with BSkyB over the name SkyNet

    1. SkippyBing

      Re: Starlink

      Or having more than one satellite network called skynet could get confusing?

  12. JimmyPage

    a recovery vessel, dubbed Mr Steven

    Surely a missed marketing opportunity there ? Either get it sponsored, or have a $1 a pop naming competition ?

    1. Mike Brown

      Re: a recovery vessel, dubbed Mr Steven

      What makes you think Mr Steven didnt already pay his pound?

    2. Ugotta B. Kiddingme

      Re: a recovery vessel, dubbed Mr Steven

      well, if they are referring to the vessel as a "catcher's mitt" then why not name it after one of the Major League Baseball Hall of Fame catchers? Johnny Bench, Yogi Bera, and Gary Carter are the three I can name off the top of my head. I know there are several others.

      1. Yet Another Anonymous coward Silver badge

        Re: a recovery vessel, dubbed Mr Steven

        Because their names are probably, trademarked, owned, licensed and litigated over.

        The advantage of using culture names for ships is that if the culture do ever contact us, the GSV will be pleased, or at least mildly amused

      2. Sorry that handle is already taken. Silver badge

        Re: a recovery vessel, dubbed Mr Steven

        if they are referring to the vessel as a "catcher's mitt" then why not name it after one of the Major League Baseball Hall of Fame catchers? Johnny Bench, Yogi Bera, and Gary Carter are the three I can name off the top of my head. I know there are several others.

        Adam Gilchrist.

    3. Milton

      Re: a recovery vessel, dubbed Mr Steven

      "Surely a missed marketing opportunity there ? Either get it sponsored, or have a $1 a pop naming competition ?"

      Maybe they did, and 97% of the first 1,000 responses was "Netty McNetface" ...

      ... at which point, the boss, emitting a long-suffering sigh, pulled up a random-name-listing-site from the 'net and said "That one" without looking at the screen.

      I'm almost surprised the business isn't called Corporation 9592.

  13. Asterix the Gaul


    If I recall correctly, Musk gave a figure of (4K?) satellites for his global internet coverage.

    I cannot understand how placing the initial few launches in areas with poor coverage would make commercial sense, as most are likely to be poorer places.

    Place them over Europe first, he will make a killing by taking on the likes of BT, Virgin, Sky & others in mainland Europe.

    I for one would happily signup.

    1. Sorry that handle is already taken. Silver badge

      Re: Starlink

      How you cover regions with satellite service depends on the orbits you put them in. Since these will be in LEO, they won't strictly be "placed" over a specific region in the way that GEO satellites are. Iridium for example uses polar orbits with the satellites in six planes. Each plane (simplistically) covers 30° each on opposing sides of the earth. The satellites orbit with a period of 101 minutes, so 11 are required in each plane to ensure that, at any point on the earth, one is roughly overhead whenever it's needed. There's a good animation of how it works on Wikipedia:

      The plan may well be to do both at the same time. If you cover western Europe with polar orbiting satellites, you'll also cover most of the western half of Africa (and the middle of the Pacific Ocean...) Similarly if you cover North America, at the same time you'll cover Asia.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Other stories you might like

Biting the hand that feeds IT © 1998–2022