Slashdot is powered by your submissions, so send in your scoop

 



Forgot your password?
typodupeerror
×
Communications Space The Internet United States

SpaceX Says 12,000 Satellites Isn't Enough, So It Might Launch Another 30,000 (arstechnica.com) 142

SpaceX is seeking permission to launch another 30,000 low-Earth orbit (LEO) satellites for its Starlink broadband network, which would be in addition to the nearly 12,000 satellites the company already has permission to launch. But it's too early in the process to determine whether SpaceX is likely to launch most or all of the additional 30,000 satellites. Ars Technica reports: The Federal Communications Commission made the requests on SpaceX's behalf, as is standard practice, in a series of filings with the International Telecommunication Union (ITU) last week. (Here's an example of one of the filings.) The 30,000 satellites would operate "at altitudes ranging from 328 kilometers to 580 kilometers," SpaceNews reported yesterday. The filings are known as coordination requests. As SpaceNews noted, the ITU coordinates spectrum "to prevent signal interference and spectrum hogging." SpaceX's filings could help the company reserve spectrum before other operators claim it, but it's an early step in the process and doesn't commit SpaceX to launching all 30,000 satellites.

SpaceX's constellation alone would dwarf the total number of satellites orbiting Earth today. As of January 2019, about 8,950 satellites had been placed into Earth orbit since 1957, and about 5,000 of those were still in space, according to the European Space Agency (ESA). Only about 1,950 of those are still functioning. If SpaceX proceeds with the additional 30,000 satellites, it would have to seek FCC permission and provide more technical detail, including plans to minimize debris and prevent collisions. SpaceX is designing its satellites to burn up completely during atmospheric re-entry in order to prevent physical harm from falling objects.

This discussion has been archived. No new comments can be posted.

SpaceX Says 12,000 Satellites Isn't Enough, So It Might Launch Another 30,000

Comments Filter:
  • But around the Earth, aren't these many things likely to collide at some point?
    • Re:Wide space (Score:5, Insightful)

      by bloodhawk ( 813939 ) on Wednesday October 16, 2019 @10:23PM (#59317070)
      unlikely. while 30,000 sounds a lot but in reality it is quite sparse. to put into perspective if you placed them on earth instead of space (obviously far less area here) then each satellite would have the equivalent of 17,000 square km's to play in. In space that number is way way larger.
      • Re:Wide space (Score:4, Insightful)

        by Ol Olsoc ( 1175323 ) on Wednesday October 16, 2019 @10:25PM (#59317080)

        unlikely. while 30,000 sounds a lot but in reality it is quite sparse. to put into perspective if you placed them on earth instead of space (obviously far less area here) then each satellite would have the equivalent of 17,000 square km's to play in. In space that number is way way larger.

        Only takes one collision to set off a chain reaction.

        • Re:Wide space (Score:4, Insightful)

          by bloodhawk ( 813939 ) on Wednesday October 16, 2019 @10:28PM (#59317084)

          unlikely. while 30,000 sounds a lot but in reality it is quite sparse. to put into perspective if you placed them on earth instead of space (obviously far less area here) then each satellite would have the equivalent of 17,000 square km's to play in. In space that number is way way larger.

          Only takes one collision to set off a chain reaction.

          no it doesn't. a chain reaction would require them to be in perfect synchronised distance from the earth and for a collision to perfectly target the next satellite over 100km away with pinpoint accuracy and for that 1 in a billion chance to happen repeatedly. I would think you are more likely to win the lotto 10 times in a row.

          • by Guspaz ( 556486 )

            When two satellites collide, it does not simply bump the satellite into a new path, both satellites shatter into a very large number of pieces moving very quickly. Those pieces then collide with other satellites, further multiplying the number of pieces.

            • Re:Wide space (Score:4, Insightful)

              by ShanghaiBill ( 739463 ) on Wednesday October 16, 2019 @11:37PM (#59317248)

              These satellites are in circular low earth orbits. If two collide, nearly all the debris either goes into a non-circular orbit which dips into the atmosphere, or does not retain enough velocity to stay in orbit.

              Even for the tiny fraction of particles that end up in circular orbits with enough velocity, they will experience significant friction in LEO, and the orbits will degrade.

              • The orbits are almost never completely circular. They're elliptical, and disturbed by LEO atmosphere and electromagnetic effects. They don't _remain_ circular. I agree that a disturbed orbit is much harder to plan for and avoid accidental overlaps.

            • LOW EARTH ORBIT. those pieces will not maintain orbit.
              • by Hodr ( 219920 )

                Depends on speeds, specific orbit, and height (LEO is a fairly big range). Some LEO satellites could stay up for hundreds of years unassisted. Others may be pulled down in a decade or less.

            • Re:Wide space (Score:4, Informative)

              by Dunbal ( 464142 ) * on Thursday October 17, 2019 @01:21AM (#59317398)

              So you're saying a collision results in a net gain of kinetic energy? You watch too many movies.

              First - satellites usually move in the same direction - from west to east. Yes there are variations of inclination and polar orbits do exist, but most satellites trace this west to east orbit. So most of the velocity component of any satellite will be in this direction. You will not be seeing "head on collisions" of any satellite. Actual collisions are likely to be much gentler than at orbital velocities as one satellite catches up and "bumps" into another one.

              Satellites are not made from high explosives and will not detonate on impact. They are satellites not warheads. Sure, deformation could happen and bits could break off, but a pair of satellites shattering into billions of pieces from a collision is a fantasy. Colliding satellites would not suddenly reverse their trajectories or instantly change altitudes - most of the resulting fragment(s) of the collision would continue along a vector that is the resultant of the addition of the velocity vectors of the original satellites. In fact - it's pretty much guaranteed that deviation from this new vector is inversely proportional to the mass of the fragment. You won't get half a satellite suddenly turning 90 degrees in a different direction, although some microscopic bits and pieces might.

              Satellites are not soap bubbles. They are hit by micrometeorites all the time and although the damage is cumulative over time, it's not likely that a single impact will completely disable a satellite. Solar flares are a much greater threat to satellites than micrometeorites are. Also - micrometeorites are likely to be traveling at much greater speeds than any micro-debris from a collision. Like I said - you are not going to be getting any head-ons. Satellites move roughly in the same direction for the component with the greatest magnitude.

              And finally: "Space is big. I mean REALLY big..."

              • First - satellites usually move in the same direction - from west to east. Yes there are variations of inclination and polar orbits do exist, but most satellites trace this west to east orbit. So most of the velocity component of any satellite will be in this direction.

                Picture two orbits 180 degrees apart, i.e. crossing the opposite poles simultaneously, ascending the equator one half "wavelength" apart, and descending it one half "wavelength" apart. Ignore the earth rotating underneath, you're in the orbits' frame of reference.
                They have to keep synchronous, so they must keep precisely at the same altitude.
                Now this would be pretty dumb to put the satellites on these orbits precisely so that they cross the equator at the same time, and there is no need to do it, but I can

              • This is a wildly optimistic view. When two satellites are 10 degrees of inclination apart, the collision speed is already 1 km/s. That's more than enough kinetic energy to shatter both into tiny fragments.

                They are hit by micrometeorites all the time

                Emphasis on the micro. Those are fragments smaller than 1 gram we're talking about.

              • It's weird that you're so confident about something not happening which has already happened in the last few years. How is this getting upvoted?

                So you're saying a collision results in a net gain of kinetic energy? You watch too many movies.

                Uhhh, have you ever seen a car accident? When you have random transmission of energy some pieces accelerate and some decelerate. As a whole there can be no net gain while half gain and half lose energy. The result is a cloud of debris: https://i2.wp.com/www.spacesaf... [wp.com]

                The iridium-Cosmos collision resulted in 1,000 pieces >4" in diameter. It was not "gentl

            • by Rei ( 128717 )

              The odds of failures propagating faster than debris deorbits is, of course, heavily modeled. The primary factor, at least as important as the number of live satellites in orbit, is the presence / reliability of their deorbiting system. So long as a satellite is alive and can maneuver, collisions are generally not a problem. Many satellites don't have a system to deorbit at all, however, and you have to wait (depending on the satellite) years, decades, or even effectively indefinitely for it to happen.

              If S

            • Sounds really cool... You should write a screenplay.....That would make a great movie.
          • unlikely. while 30,000 sounds a lot but in reality it is quite sparse. to put into perspective if you placed them on earth instead of space (obviously far less area here) then each satellite would have the equivalent of 17,000 square km's to play in. In space that number is way way larger.

            Only takes one collision to set off a chain reaction.

            no it doesn't. a chain reaction would require them to be in perfect synchronised distance from the earth and for a collision to perfectly target the next satellite over 100km away with pinpoint accuracy and for that 1 in a billion chance to happen repeatedly. I would think you are more likely to win the lotto 10 times in a row.

            Go learn some orbital mechanics and what happens to space debris. Tell us what happens to devices that manage to turn into various pieces of space junk. Which pieces are accellerated, and which are decelerated. Your offhand claims of odds just exposes that you don't know as much as you think you do.

            Although if you wish to provide the math - I'll listen.

          • Hmph.

            LEO Satellite Zero is struck by a discredited* projectile. Debris is scattered above, below, and along previous orbital path. Spreading, impacts on neighboring satellites occur. Each increases the rate and number of impacts. Some move the targets more than they can recover from, new orbits, unexpected paths, other satellites are impacted well outside of the intended paths.

            You can't control this chaos. Hilarity ensues.

            Dammit, I cannot win the lottery soon enough, to bankroll a space garbage truck projec

        • by Dunbal ( 464142 ) *
          Anti-logic. If they are so sparse that a collision is unlikely then a collision setting off a chain reaction (requiring a second collision) is even LESS likely. You get chain reactions when collisions are likely not when collisions are unlikely.
          • Faux-intelligence. (aka The Slashdot Special). Collisions result in debris. https://i2.wp.com/www.spacesaf... [wp.com]

            Two satellites colliding is improbable. 1,000 pieces of debris >4" like the Iridium-Cosmos accident means 1,000x more chances of further collisions. And while those 1,000 pieces may not be big enough to completely destroy a satellite and create 1,000 more pieces alone, they are enough to disable a satellite and make it unable to maneuver away from other satellites or more debris.

      • by 4im ( 181450 )

        Sure, space may be huge. But what about communications with these satellites? Much of the spectrum is reserved for different tasks. Is there going to be enough available to properly control this many sats? Especially if competitors want to do the same thing.

        As an amateur astrophotographer, I'm not looking forward to these sat constellations. Sigma stacking may alleviate some of the issue, but still...

        And all this trouble only because some countries (even supposedly first-world) can't properly handle resourc

      • by AmiMoJo ( 196126 )

        The bigger issue is the amount of launches required. SpaceX is putting up 60 satellites in one go, which to maintain a constellation of 42,000 with a operating lifetime of 5 years each would require about 3 launches per week indefinitely.

        Of course any unexpected failures may result in them needing extra launches to deal with them.

        Even assuming they can get the cost down far enough that's a lot of pollution to be putting out.

        • SpaceX is putting up 60 satellites in one go, which to maintain a constellation of 42,000 with a operating lifetime of 5 years each would require about 3 launches per week indefinitely.

          That has bothered me too. OTOH, I think Space X used the Falcon 9 Block 5 for that launch which has a payload of ~50,000 pounds. If they used a Falcon Heavy with a payload of 150,000 pounds, could get that launch schedule down to 1 launch per week. It still doesn't seem doable but I suppose that depends on their subscription profits.

          • by AmiMoJo ( 196126 )

            Even one a week is a lot, and 42,000 disposable satellites every 5 years is a lot of material.

            I'm kinda surprised that the economics work out for it. Maybe it's a loss leader to get their overall launch cost down by creating volume.

            • I'm kinda surprised that the economics work out for it. Maybe it's a loss leader to get their overall launch cost down by creating volume.

              I'll try to make the numbers work, but my source selections may be FoS. In this thread ...
              https://www.reddit.com/r/Starl... [reddit.com]
              Posters suggests the throughput for each satellite is 20 Gbps, so 1200 Gbps for each launch of 60.

              Assuming they can spread that 1200 Gbps out to 9600 subscribers of 1Gbps service (is that too liberal? Conservative?) at $50 a month for five years yields ...

              $50 × 12mo × 5yr × 9600clients = $28.8 million in revenue per launch.

              According to ... https://www.googl [google.com]

              • A falcon 9 launch doesn't cost anywhere close to $62M. That's the base price SpaceX charges for one, but it includes a very healthy profit margin to do things like recoup investment, fund future investment, and ensure sufficient assets to survive unexpected losses. The actual cost is far, far cheaper - the exact numbers aren't publicly shared as far as I know, but the estimates I've seen run somewhere around $28M for a new vehicle, and under $15M for a re-flown one (the majority of which is the always-new s

            • by Rei ( 128717 )

              The global internet access market is over $1T per year, and growing, with large swaths of the planet unconnected or underconnected. Even taking a small fraction of the market is worth it.

              • by AmiMoJo ( 196126 )

                Hmm, but I wonder how competitive it will be. The parts where there is no cabled access are not going to pay much for internet access, and the parts where there is cabled access are going to vastly out-strip it for speed. Also you will need equipment to access it, where as most people already have a phone line at least into their house.

                Maybe it will be worth it but it really feels like a scheme to get launch costs down.

                • Why would you expect cabled access to be faster than satellite? There's no physical reason that needs to be true. Both the satellites and the user terminals use phased arrays to generate quite narrow beams, so there's not that much competition for bandwidth, and with so many satellites they can manage quite staggering throughput. Existing satellite service is slow (in different ways) for reasons that don't apply to Starlink: low-orbit constellations mostly don't use highly directional antennas (hard when yo

          • by Rei ( 128717 )

            SpaceX considers both Falcon 9 and Falcon Heavy obsolete. They're switching to Starship, with a ~150t fully reusable payload to LEO.

            More to the point, SpaceX wants to justify frequent launches. They want to have a reason to keep their rockets busy. :)

            • A friend of mine who works at SpaceX has joked that the real reason Musk wants Starship/SuperHeavy is because it could put 400 Starlink birds up with a single launch.

              Mind you, even at the incredibly fast rate Musk is pushing Starship, I expect Starlink will begin paid service before the first time Starship puts any payload into orbit.

        • Of course any unexpected failures may result in them needing extra launches to deal with them.

          Three of the initial 60 have already failed, but I think that is mostly growing pains.
          https://www.businessinsider.co... [businessinsider.com]

          Even assuming they can get the cost down far enough that's a lot of pollution to be putting out.

          I think the only concerning pollution will be to stargazers. I think the benefits outweigh those costs. That's just me.

          • Working out the numbers with my wee brain, I come up with each Falcon heavy launch using the fuel of roughly 30 tank-loads of a Boeing 737-sized airliner.
            https://en.wikipedia.org/wiki/... [wikipedia.org]
            https://www.quora.com/On-a-Boe... [quora.com]
            That seems like a lot but according to...
            https://www.google.com/search?... [google.com]
            on average, there are ~10000 of these size airliners in the air at any given moment. So in terms of air pollution, it's a gnat's fart in a hurricane.
          • by Rei ( 128717 )

            Even a lot of people trying to deliberately spot them with the naked eye failed, esp. after they spread out. And SpaceX plans to take more measures to reduce their visibility. I don't expect people in general to notice a change to the night sky. Ground-based telescopes of course will, but as launch costs continue to fall, we'll be seeing more of a push to space-based telescopes (also, ground-based telescopes already use algorithms to remove satellite trails from their datasets)

            In terms of night sky pollu

        • Iâ(TM)m going to guess the plan is to use their starship to do it. In the latest presentation Elon said that if ALL the rocket companies on Earth, including spacex, launched at their maximum rate they could put 300 tons in orbit a year. Four starship launching and relaunching could put up ONE MILLION ton a year. Even if we account for his crazy optimism and say its 1/8th of the planned amount, its a crazy quantity. Read up on starship, the pace of progress on this platform and the amount of cargo space
        • They'll be going up mostly at 7x that rate on Starship/Heavy, both of which run on methane which will be "farmed" at the launchsite from atmospheric CO2, seawater, and Tesla solar cells.

          Each Starship/Heavy stack can be launched three times a day and they've currently building four of them.

      • It doesn't work like that. These satellites aren't stationary. Each satellite is circling in an orbit, and you have to prevent those orbits from intersecting Preventing intersections at uneven intervals requires putting them all in the same orbit, spaced out. Or at different altitudes if they're high enough that the orbit won't decay appreciably from atmospheric drag. So 30,000 satellites in a few orbits at different altitudes and inclinations would result in the satellites being maybe a dozen km from e
      • each satellite would have the equivalent of 17,000 square km's to play in.

        Yeah, but since they are traveling at 7.8 km/s, they are covering that 17,000 sq km quickly...

    • by AHuxley ( 892839 )
      They are low and lots of space around them all.
      Gets the ping down and ensures everyone gets some much needed bandwidth.
      The alternative has been used for years. Got "a" huge sat and try that high above the earth position.
      Limited bandwidth is shared by many more users, costs are up and ping is huge.
    • by Jeremi ( 14640 )

      One has to assume that SpaceX isn't just going to be throwing them into random positions and velocities, but rather they will be guided into carefully selected orbits such that they are evenly spaced and unlikely to collide with each other.

      No doubt that will be something of a challenge, but SpaceX has solved much harder problems than that already, so I'm confident they can meet it.

    • I sure hope so
  • by timeOday ( 582209 ) on Wednesday October 16, 2019 @10:16PM (#59317056)
    This reminds me of the early days of the Internet when the more prominent universities and companies were handed class A IP4 addresses (that is, each got 1 / 256 of the entire IP4 address space to themselves).

    This could be OK but it should only be granted on a "use it or lose it before year 2030" or whatever basis. Perpetual ownership of any limited natural resource is actually wrongheaded.

    • Bear in mind that in the early days, the "Internet" was supposed to be a Defense Department communications network [wikipedia.org]. Nobody knew it was going to become The global computer network. And when they realized that was happening, they developed IPv6 to replace IPv4. Except nobody adopted it.
    • by AmiMoJo ( 196126 )

      It's a bit complicated. Once granted you have 7 years to start using the frequency, but even then you have to work with governments and other companies to make it inter-operable. You can't just camp the frequency.

    • The issue you raised about IPv4 was recognized and resolved over two decades ago through the creation of a less provincial, more far reaching standard. That isn't going to work on the EM spectrum.

      You seem to be confusing limited with scarce. EM spectrum is relatively scarce, yes, but it is not limited. There is a huge difference, legally, and more importantly, economically, between "limited" and "scarce." Unlike oil and other finite natural resources, the EM spectrum is going to be exploitable for a long

  • Can't leave my house anymore without my aluminium hat. =/

  • Wow. That's a lot of satellites and a lot of launches. If they have a 5-year lifespan, it will still require 8400 satellites a year to maintain the system. Assuming 60 satellites per Falcon 9 launch, that's 140 launches per year ( almost 12 per month or 3 per week).

    edit: the above numbers are loosely based upon a constellation total of 42,000 satellites, not just the new 30,000.

    Starship should be able to launch around 300 at a time. So figure on 28 annual launches just for Starlink. It's still an impressiv

    • Wow. That's a lot of satellites and a lot of launches. If they have a 5-year lifespan, it will still require 8400 satellites a year to maintain the system. Assuming 60 satellites per Falcon 9 launch, that's 140 launches per year ( almost 12 per month or 3 per week).

      edit: the above numbers are loosely based upon a constellation total of 42,000 satellites, not just the new 30,000.

      Starship should be able to launch around 300 at a time. So figure on 28 annual launches just for Starlink. It's still an impressive number, especially considering the production volume for the birds themselves.

      Falcon 9 would indeed be untenable as the launch system for that many satellites in LEO. But SpaceX has already publicly stated they intend to use Starship Cargo for Starlink launches. They will be able to put up enormous numbers in a single launch. Starship's ballpark target cargo capacity is 100,000 kg, which they hope to improve to 150,000 kg in later revisions. Falcon 9 can put 16,800 kg into LEO when the first stage is reused. Starship should start out being able to put a factor of 6 more Starlink

  • basically these look to be a network of powered reflectors (steered) thats nice I hope they are building into it some sort of positioning and if so it looks like GPS superiority will end...

  • " it would have to seek FCC permission "

    Those satellites will be orbiting the whole planet, not just the USA.
    Won't the other (space capable ) countries object?

    • FCC permission is required to broadcast on frequencies reserved for communicating with satellites. So if your guidance and control satellite dishes are in the U.S., then you need permission from the FCC. If you want to place those control dishes in other countries, then you need to get permission there. If you wanted to set up such a communications station aboard a ship in international waters, then you shouldn't need any permission. (These frequencies are reserved because satellites broadcast with very
  • by sad_ ( 7868 )

    we really have a hard time cleaning up after ourselves, don't we?

    "As of January 2019, about 8,950 satellites had been placed into Earth orbit since 1957, and about 5,000 of those were still in space ... Only about 1,950 of those are still functioning."

    more than half of the satellites are just waste, garbage, polution. i guess nobody at the time even had the smallest though what to do with them when they reached end-of-life (i though the general concencus was - burn them up - but this is clearly not the case

    • A lot of those satellites are orbiting too far from earth for re-entry to be feasible. Once you're in a high enough orbit, it takes quite a lot of delta-v (change of velocity, essentially "maneuvering") to dip your orbit low enough to re-enter on any useful time scale. End-of-life geostationary satellites are actually usually boosted into a higher orbit (called a "graveyard orbit") that is far easier to reach than the atmosphere.

  • How long until the Starlink broadband network turns into a giant waste of time and money---like the Hyperloop tunnel which is just a regular tunnel now.

  • And with a large number of them needing a clear frequency without interfering with other satellites where are they going to get the required spectrum?

    • And with a large number of them needing a clear frequency without interfering with other satellites where are they going to get the required spectrum?

      You could just RTFM [fcc.gov].

      They don't interfere with each other. They're in low Earth orbit, so low that the majority of satellites don't even have line of sight with each other. Those that do aren't transmitting at each other. They're using phased array antennas to talk to quite a small patch on the ground.

      Where are they going to get the required spectrum? From the FCC.

      10.7-12.7 GHz (space-to-Earth), 12.75-13.25 GHz (Earth-to-space), 13.85-14.5 GHz (Earth-to-space), 17.8-18.6 GHz (space-to-Earth), 18.8-19.3 GHz (space-to-Earth), 19.7-20.2 GHz (space-to-Earth), 27.5-29.1 GHz (Earth-to-space), and 29.5-30 GHz bands (Earth-to-space).

      There you go.

  • Whats the average round trip time for these bad boys? They don't suffer from refraction pushing through glass right?

A morsel of genuine history is a thing so rare as to be always valuable. -- Thomas Jefferson

Working...