From what I’ve read, these things are not much bigger than a home laser printer. They obviously have nowhere near enough power to be dangerous as microwave sources to humans or other satellites. As for physically targeting other satellites, they have those ion engines that are remarkably low-powered. I presume it would take days, weeks, possibly months to move into a collision course with other satellites.
I assume nowadays anyone producing a device that sits somewhere waiting for instructions have programmed decent security. (This assumption possibly may not be valid). Regardless, I have not heard of satellites being compromised, and assume Starlink is no different. Plus as I understand, they can be reprogrammed if necessary
They’re much larger than that–about 260 kg each. More like the size of a large table than a laser printer. Unfolded, they’re about 10 meters high (almost all of that is solar panel). Interestingly, they can re-fold their panel to reduce collision risk (by reducing the effective cross-section):
They’re still relatively small compared to most satellites, but nowhere close to as small as cubesats.
The starlink I have at our cabin is so portable and incredibly easy to setup, turn off and repeat. I can imagine how incredibly useful it could be in a warzone or anywhere if you need to be mobile. You just need a power source and you can have the high speed internet router going in just a couple of minutes. Unplugging and carrying it to another location would be easily doable for just 1 person.
Huh, dang. Direct satellite-to-cell connectivity. T-Mobile giving a chunk of their existing spectrum to Starlink. Existing T-Mobile phones will connect to Starlink. Starts with text/MMS but later voice/MMS. Even global texting is pretty amazing.
2-4 megabits/sec per cell zone. Sounds low at first, but that’s really a huge amount. That’s enough for basically unlimited texting, and a thousand or so simultaneous voice calls. Since you’ll only be using this in super out of the way areas, that’s more than enough.
I was not expecting this at all–best guesses about the announcement were some kind of remote cell tower system, maybe with a normal Starlink antenna and some solar panels. I did not consider a direct-to-phone connection remotely plausible, let alone one that works with existing phones.
It requires Starlink v2 satellites, has an extra dedicated antenna (~5x5 meters!), and the satellites handle all the Doppler shift and other required elements. It won’t require full deployment for the v2 satellites since texting is basically asynchronous. They’ll make it work with other messaging apps as long as they can guarantee that they’re sending texts (i.e., low bandwidth stuff).
Egads!! No, not expecting that. International rollout is going to be interesting. Lots of vested interests, but for countries with big remote areas this is a seriously big thing. Not to mention on the oceans.
There are a few existing players in satcoms that will be very worried.
Starlink might think about AIS and Epirb receivers to address seafarers. Likely a very tiny addition.
Ok, you might be asking: how does this even work, a tiny phone talking with a satellite? Aren’t they hundreds of miles up? What’s the catch?
Any wireless communication system can be boiled down to a few basic numbers:
The distance between the two antennas
The product of the area of the two antennas
The amount of power sent through the antennas
The desired bitrate
The size of the frequency band (the original meaning of “bandwidth”, though nowadays it tends to be used for bitrate)
That’s an approximation of course, but it’ll do for now. The reason it works is that first, they’re targeting very low bitrates: texting and maybe voice. Not video calls or 5G-speed downloads.
And although the phone has a tiny, low-power antenna, the satellite antenna can be huge. And indeed, they’re targeting a >250 square foot antenna on the satellite. That big antenna can gather the weak signal from the phone, and can also be made very directional, like a spotlight with a big reflector.
The satellites can pump out a lot of power, too. More energy equals more bits.
There are some difficulties. Doppler shifting is one–the fact that the satellites are going so fast that the frequency is raised as its traveling toward you, and decreased as it’s going away, like the sound of an ambulance as it passes. The phones likely have minimal Doppler correction (since it’s only designed to work at low speed), so the satellite will have to handle it. But that’s largely a software problem, assuming the hardware was built with enough flexibility.
Agreed. A while back I looked into Iridium SBD (short burst data). As I recall, the pricing worked out to be something like $1 per kilobyte. The receiver was impressive; about the size of a matchbox and needed a 1" square patch antenna, but the prices were too high for anything but some very simple telemetry. This Starlink thing isn’t exactly broadband, but could easily be 1/100 the price per byte.
Musk had this to say earlier about transmitting into a country without authorization:
It’s an uncensored link. Iran is of course not approving this. However, it was not a unilateral decision by Musk; he had approval and encouragement from the US State Department. In fact it seems possible that they came to him first, given the protests there.
@thorny_locust : You seemed to be asking in the other thread about the risks Starlink presents to spaceflight in general, and presumably related to space junk mitigation efforts.
The short and sweet version is that Starlink aims for a 5-year maximum time in orbit after the end of useful activity. They accomplish this in a few different ways:
The satellites are at a low altitude and experience a fair amount of atmospheric drag. The amount of drag varies based on solar activity, but within a 5-year window there is enough to bring the satellites down.
The satellites have a low ballistic coefficient, which is just a fancy way of saying that they have a giant sail on them. There’s a lot of surface area compared to mass.
They have thrusters which can actively bring the craft out of orbit
They have reaction wheels which can orient the satellite to present the maximum surface area
They are initially flown to orbit at an even lower altitude, requiring the thrusters bring them up to the operational altitude
With the active measures enabled, the satellites can be disposed of in months. With partial measures, some small number of years. And even with a completely dead and unresponsive satellite, they are estimating 5 years.
Note that 5 years is the new target for the FCC (yet to be fully enacted):
The previous rule was a 25-year life. I think the new rule is the right thing to do, and also that SpaceX should be commended to building to that new standard even before it was a requirement.
Thanks. That’s useful to know. At least there is some proposal for how to clean up the mess.
Do you know whether the number of Starlink satellites proposed to be active at any one time, presuming full deployment, would itself be a problem even if the inactive ones are guaranteed to be gone within a few years?
The answer to that depends on how quickly we can upgrade the infrastructure for tracking satellites and potential collisions.
The raw number of satellites isn’t an inherent danger. Space is big. But it’s not so big that there’s zero chance of collision–there have been a few before. So if we’re going to massively increase the number of satellites out there, and have multiple mega-constellations, we need some central tracking authority; an equivalent of an air traffic control.
The Starlink satellites have their own collision avoidance systems, but it’s just their own thing. Right now, potential collisions are resolved mostly by email, which is kinda silly. And not everyone is on the same page here; every party has their own tolerance for a probability of collision, so there are cases where one party has a fit while the other says no problem.
There are cost questions, too. If two satellites are likely to collide, which one has to move? The propellant isn’t free (using it reduces the lifetime of the satellite). Ideally, there would be a sort of bidding process where they share the costs, but only the satellite with the lowest cost makes the maneuver.
I don’t know what the US military is currently using in Afghanistan, etc. However, I do know that Starlink is far superior, to the point that they are concerned about its proprietary nature:
Without mentioning Starlink by name, Shimmin said “there is a particular commercial satellite provider in low Earth orbit, for example, that is several years ahead of the rest of the field. They do tend to try to establish interfaces, proprietary interfaces, that don’t necessarily want to play well with others.”
I hope SpaceX does address the interoperability issues. But the DoD wouldn’t be worried about this if Starlink were just one of several roughly equivalent providers. They really are years ahead.
Not just years ahead, but their lead will continue to grow. No one else can match the launch rate SpaceX is already managing. Amazon using ULA’s disposable rockets will be far more expensive and launch at a tiny fraction of the rate of SpaceX.
If Starship works as promised, no one will touch SpaceX’s ability to put mass in orbit for at least a decade and probably much more, assuming they start building Starship competitors very soon after it demonstrates its capability. And by then, SpaceX will be a decade ahead in placing its LEO constellation.
I suspect SpaceX will eventually sell launch services to its Starlink competitors, if for no other reason than to forestall monopoly accusations and litigation.
They needed a new launch provider after–oops–Russia invaded Ukraine; Russia said we won’t launch your shit on Soyuz unless the British government pulls out; OneWeb says no; Russia says ok, no launch, plus we’re not giving back the satellites you shipped to us.
I see no evidence that SpaceX won’t sell their launch services to anyone at their standard price.
Here’s a gift article from today’s Washington Post. It mostly talks about the train of satellites and impacts on astronomy, but also potential issues with failures that may interfere with space stations. Starlink satellite trains
There is an article today about how one of Canada’s big telcos has signed a deal with Musk. There are huge areas with poor connectivity, so perhaps on balance it is a good thing for them. I don’t know exactly what it will mean in practice.
The heliski lodge I go to in northern BC got Starlink this year, and it is kind of a huge deal. They were previously on Telus and the service was very expensive (like over $10k/month) and pretty crappy. 20 guests and 20 staff using 3 Starlink units, no problem.
