Slashdot Mirror
SpaceX Files FCC Application For Internet Access Network With 4,425 Satellites (geekwire.com)
An anonymous reader quotes a report from GeekWire: SpaceX has laid out further details about a 4,425-satellite communications network that's expected to provide global broadband internet access, with its Seattle-area office playing a key role in its development. The plan is explained in an application and supporting documents filed on Tuesday with the Federal Communications Commission. In the technical information that accompanied its application, SpaceX said it would start commercial broadband service with 800 satellites. That service would cover areas of the globe from 15 degrees north to 60 degrees north, and from 15 degrees south to 60 degrees south. That leaves out some portions of Alaska, which would require a temporary waiver from the FCC. Eventually, the network would grow to 4,425 satellites, transmitting in the Ku and Ka frequency bands. "Once fully deployed, the SpaceX system will pass over virtually all parts of the Earth's surface and therefore, in principle, have the ability to provide ubiquitous global service," SpaceX said. The satellites would orbit the planet at altitudes ranging from 714 to 823 miles (1,150 to 1,325 kilometers) -- well above the International Space Station, but well below geostationary satellites. SpaceX said it would follow federal guidelines to mitigate orbital debris. Each satellite would weigh 850 pounds (386 kilograms) and measure 13 by 6 by 4 feet (4 by 1.8 by 1.2 meters), plus solar arrays, SpaceX said. Operating lifetime was estimated at five to seven years per satellite.
It has a fine series of tubes for the intarwebs!
With this move, won't SpaceX be competing with their own clients like Iridium?
... that'll be another 4425 bits of space junk. Genius idea - utterly pollute near space just so some company can make a short term profit on something thats a nice to have rather than essential infrastructure.
"SpaceX said it would follow federal guidelines to mitigate orbital debris"
And how does it plan to do that exactly? They're too high to be sent down to burn up in the atmosphere and too low to be sent off into a parking orbit.
Rural Africans etc. will install satellite tracking dishes and there'll be a half minute interruption every few minutes as satellites fly by? Or it doesn't need line of sight?
Wow, even assuming you had no need to lift fuel for the last mile those preceding hundreds of miles, and you didn't need to put the satellites in rockets or anything, I get an amount of energy that seems ludicrous. Like power Chicago for a year kind of crazy.
Nice to know we won't slow down boiling the earth until after everybody can read about how we did it to ourselves.
...there isn't a secondary payload of 4,425 RNM satellites too...
Considering that there are currently only about 1000 satellites in operation on orbit that figure of nearly 4500 sounds an awful lot.
The hype must go on.
2/3rds of the satellites will always be over water and have their bandwidth utterly wasted. A significant part of the rest will be over areas where almost nobody lives, or nobody can afford to pay for internet with hard currency. Meanwhile all 400m Europeans that live in the populated 5m square kilometers have to use the same 20 to 100 satellites.
Because the satellites are not geostationary they'll need to use omnidirectional antennae which puts some hard limits on bandwidth, while a lot of people will get FTTH and 5G mobile networks in the next decade.
Iridium can get away with these shortcomings because they target the customers that doesn't care about prices. But I kind of doubt that market can support 4000 satellites
What kind of bandwidth / latency does that translate into?
Slashdot: where don knuth is an idiot because he cant grasp the awesome power of php
The math may be off; take it for what it is worth. Making the assumption that the final rollout would cover 1/2 the surface area of the earth (sounds like they are trying for more) at the minimum altitude given, that would be one satellite per ~80,000 Sq KM. add in all the different altitudes you could put these at (another assumption of ~15 Km of altitude difference which seems generous) you are looking at satellites at the same altitude being ~1000 KM apart from each other. Although with lifespans of 5 to 7 years. I still think it would be a good idea to provide the capacity to deorbit them, or 20 years in things may be getting pretty hectic. IANARS, but I've played a lot of KSP.
That is an enormous amount of weight to send up. Space-x is aiming for (has not achieved) $1,000 per pound. Their current cost is more realistically $4,000.
4425*850*4000=$150,450,000,000. Then add the cost to send up another 4427/7=630 satellites per year (630*850*2000(because they'll get costs way down if they can send up that much material)=$1 billion dollars per year. They need to spend 150 billion dollars initially and an ongoing 1 billion dollars per year.
In 2014 SpaceX had a "market cap" of (optimistically) 12 billion dollars. Let's assumt that 12 billion dollars have already been justified. Now rumors of an IPO have been heard, so let's assume a massive over-the-top IPO: 13 billion dollars. Then add in a billion dollars. (assuming every penny they can scrape together goes to this plan) 12+13+1=26 billion. Using realistic numbers for launch costs and hyper-optimistic numbers for funding, they're about 125 billion dollars short. And I don't see Trump signing a 125 billion dollar Space-X pork bill. If we're very optimistic about launch costs that hypothetical bill could go as low as a still-highly-unlikely 75 billion dollars.
If video games influenced behavior the Pac Man generation would be eating pills and running away from their problems.
Musk touts his green credintials with Tesla, yet then he proposes something like this. Hypocrite of the first order.
386 kilograms - 13 by 6 by 4 feet (4 by 1.8 by 1.2 meters)
Launch cost $1.79 million per satellite
or $7.93 billion per constellation which last 5-7 years.
Lets say an average cost of $50 million per satellite, which is very low, were looking at $221.3 billion.
So, let's say roughly $230 billion, just to break even over a 7 year life span would require an annual revenue generation of $32 billion. Given that most people without an internet connection would be in rural areas, or poverty striken areas, we're looking maybe $30 per month or $360 per year. So, that would require a user base of 91 million people requiring each satellite to host 20000 people.
The numbers here don't seem to make sense.
Unless I'm being stupid (of which there is a high possibility) then if there are 4425 satellites and they last 5 years each, that means every year 885 of them will have to be replaced. That equates to between 2 and 3 per day. Assuming each one requires a separate launch, is it really viable to have that many trips to space every single day for the rest of time? Does the business model support it? Whilst space exploration is one of the things that most appeals to my sense of wonder, I have to admit that to me this seems like an awful lot of fuel that could be spent on other cool space things (like journeys to the moon, or mars, or to Europa). But then again I'm not crazy old Professor Hyperloop so what do I know?!
More junk ? Come on SpaceX! If it was something innovative, like a broadband network to the moon where a moon based relay transceiver could be used to communicated with satellites, probes, rovers on Mars and beyond, that would be cool. Data rates will be faster. No terrestrial background noise to deal with.
The transmitted signal from the moon can be extremely high, and powered by a small nuclear reactor. But a broadband network consisting of thousands of tiny satellites ? Not a good idea.
Why think small... can't we learn from Iridium ? No money made there. Plus why litter Earth's orbital space with debris ?
7 year max lifespan, times 365 days a year, is 2555 days of maximum lifespan.
So for 4400 satellites, how exactly are they going to maintain a launch rate capable of sustaining this? This would inevitably require maintenance launches 2 or 3 times per week.
Who is this that even the wind and the waves obey Him? Surely this computer must submit also!
Such an endeavour would shake up China’s censorship of the internet (or anyone else for that matter). There’s a strong potential here for Internet 2.0 — as independent of terrestrial WAN, while "Free" in the sense that might operate independent from the regulations of any one country.
SpaceX or not, the number of satellites in orbit is going to significantly increase. Today's satellites are far lighter and more capable than before, demand is higher than ever before, and launch prices are falling across the board. Also, it's critical to note that LEO satellites like this are generally much smaller and cheaper to launch than GEO satellites.
It's times like this I wish I had a friend named 'The Professor'.
Government, ngo's, rich folks, business, or low latency traders seem obvious customers.
But these are not at the scale of the system proposed.
Here's their tech specs filed with the FCC.
http://cdn.geekwire.com/wp-content/uploads/2016/11/Technical-Attachment.pdf
They will provide service with phased array antennas.
The space terminals will cover the earth's surface with hexagonal cells.
Satellite altitude is 1150km, beamwidth is 2degrees, which make each cell 40 km across or 240kmkm.
Assuming the math is correct, it would take 620k cells to cover all land or 2.1m cells for the whole Earth.
Quite an ambitious project.
To meet interference requirements and to reuse frequencies between cells, they are using phased array antennas at the user terminals on the ground.
This probably says that the service is not for handsets?
Perhaps the roof of a boat, vehicle or building?
Perhaps a slightly different market than the current cell network.
More bandwidth, worldwide coverage, but bigger user stations.
The stated purpose is to fund the Mars stuff.
Seems likely to work for that.
Too soon to tell if the funding source is the network investors, or actual customers.
I would not be surprised to see FH launch these originally, and then the second round is done with several BFR.
We in fact desperately need an orbital based internet to break the earth bound telecommunication firms. Expect Cable companies to lobby heavily against this.
Each satellite should have a system to retrieve it to earth. By that, I mean a small system that would be deployed upon failure to push the satellite down to lower earth orbit until it plunges and burns up in atmosphere. This should in fact be a requirement of EVERY satellite.
I wonder if this will break the back of the tethered-wire monopolies. I wonder if this will give SpaceX a dominant presence in the "can you hear me now" space.
Probably not. Those folks live (and continue to live) on backroom deals and continuously decreasing value to the customer for higher and higher prices.
-engr.student
"Operating lifetime was estimated at five to seven years per satellite."
Doesn't that seem like a really sort span of time to have to send something into space? That means in like any given year you could be replacing 20% of your satellites? I guess perhaps with the idea that technology would be advancing so a 30 year old satellite might not really support current technology... Anyway still seems a bit crazy...
Tesla uses internet to update and monitor their vehicles.
This would give Tesla access around the globe to perform those updates to any Tesla, no matter WHERE in the world it is.
And they would no longer have to pay telecoms fees to use their cellular systems.
22 million subscribers @$60/month or $720/year for lowest tier of internet.
$14,400,000,000/year in revenue.
100+ million broadband users in the U.S.
220+ million cell phone users in the U.S.
Let's theorize a $50/month fee for internet and cell service. That's
$600/year. Let's say these combine to 200 million users.
That's $10,000,000,000 ($10 billion/year)
What if Space X incorporates these so called satellites into EVERY rocket launch. Think about it. They send up a rocket, it is launched, encloses a commercial payload. A cap is opened up, it is released. Done.
Wait, we just sent up a cap, and an enclosure. Could we modify our cap to double as a signal dish. Can our enclosure double as solar panels. Can we turn every rocket into an internet providing satellite?
This just proves that SpaceX will be a short-lived company. I doubt they have the capital to put it in place, let alone maintain it. And for what?
I thought satellite Internet was one of the worst options available? Most expensive, least performance?
"That service would cover areas of the globe from 15 degrees north to 60 degrees north, and from 15 degrees south to 60 degrees south." - where is the +- 15 degrees around equator exclusion zone coming from?
Link this to AI and we got something. Time to invest in our new robot overlords. Can't be any worse than Trump.