Interesting there is an optical networking option for end users (claims ~6TBps). Maybe a really dumb question, but how would the end user's ground station maintain connectivity during cloudy weather? Do they have cloud-penetrating lasers from the MEO satellites? Would that interfere with aircraft, astronomy tools, etc?
Some short googling says they have lasers that clear a path for a data carrying beam, but that seems wasteful/infeasible for commercial uses
Some info from NASA optical communication page.
"Even Earth’s atmosphere interferes with optical communications. Clouds and mist can interrupt a laser. A solution to this is building multiple ground stations, which are telescopes on Earth that receive infrared waves. If it’s cloudy at one station, the waves can be redirected to a different ground station. With more ground stations, the network can be more flexible during bad weather. SCaN is also investigating multiple approaches, like Delay/Disruption Tolerant Networking and satellite arrays to help deal with challenges derived from atmospheric means."
https://www.nasa.gov/technology/space-comms/optical-communic...
Some more info on Optical Communications for Satellites: https://www.kiss.caltech.edu/workshops/optcomm/presentations...
Seems like reusing some of Star Wars research could be used as well where the beam is constantly adjusted with independent mirrors to keep the beam coherent through the atmosphere. Also learned was the beam itself starts to distort the atmosphere requiring even more adjustments.
Wouldn't the angle of the offset matter? It seems like it would make scattering worse to be off-axis by too far.
Which then also means you have to build ground stations in this range yet far enough apart that they experience different weather yet close enough that you can redundantly link all the sites.
Aside from government and massive telecommunications companies who would this serve?
???
It's just really cool sci-fi tech that I want to see used in something other than DLP chips!
JWST and other observatories with segmented primary mirrors kind of use the segment alignment one time to get the correct alignment once. Then there is Adaptive Optics. It's kind of the opposite direction though as they are using a laser to detect the distortion so it can be compensated in the image. From learning about SDI when I was a kid/teen, it's just always been about controlling the laser itself in my mind.
The JWST does not have to deal with atmosphere or weather and uses a giant sun shield to keep the internal temperature stable so these alignments have the longevity you need to make the platform work.
Yes, maybe my comment wasn't clear if you're thinking I thought JWST was using AO. It used segment control for alignment once.
>the beam itself starts to distort the atmosphere requiring even more adjustments.
may be something like this - a high-power impulse making a channel through whatever clouds, mist, dust and after that information carrying ray/impulse through the channel, rinse and repeat
Multiple ground stations are a fine solution for backhaul, but not most end-user use cases.
So it'll never work in Scotland.
9600 Baud...
CONNECT 9600
With both RF and optical you could see FEC or ARQ being used for something that isn't 100% signal loss. Downlink is optical, uplink is RF. Downlink transmits with FEC, user terminal fixes as many errors as possible, still missing packets so requests ARQ and either gets retransmission on optical or RF.
I think customer speeds is 144 and the 6Tb is their ground links to their stations. That is my take on it at least as its not super clear. I'm curious as to how it works as well.
My read was that they're going to have 144 Gbps RF for both regular users and their ground station gateways, and 6 Tbps optical for satellite-satellite back haul, but then you can also buy direct ground-MEO access to a back haul link. (Presumably MEO-only because it's hard to maintain the link to a fast-moving LEO satellite?)
They don't seem to mention using optical for their own ground stations - maybe too unreliable?
Looking forward to TeraWave. We need a minimum of two in critical services. Google and Microsoft and Apple. Anthropic and OpenAI and Gemini.
Assuming all these companies are interested in launching their own constellations of ~10K-100K satellites into L/MEO, how many companies could actually do this before cascading collisions starts becoming a real worry?
> how many companies could actually do this before cascading collisions starts becoming a real worry?
Twenty of them at 100,000 birds each to start approaching the density of planes in the sky [1]. Not around an airport. In all of the sky. Oceans and all.
Practically speaking, this is not a pressing concern for our generation.
It's interesting that people have a hard time visualizing this. The area in Earth's LEO is, definitionally, bigger than the Earth itself.
The SEA parking garage fits 12,000 cars in it. Two of those spread over the entire planet would be an imperceptible amount of space. You could drop a pin on a map your entire life and probably never hit one.
SEA parking garage? Unfamiliar with this size reference.
Here, "SEA" = "Seattle Tacoma International Airport" in the state of Washington, USA.
That is ... an oddly specific reference?
It's one of the largest parking garages in the world
This is fair but the cars are stationary, occupying a parking space. Satellites occupy a ring.
Speed matters a lot. You can fit a lot more walking people than speeding motorcycles in the same space.
Satellites need to travel at 8 km/s to not fall down.
> Speed matters a lot
Not really. You're correct inasmuch as it increases collision energies. But it also increases momentum, which maintains orbital integrity within predictable bounds. Nobody is maneuvering around satellites, they–and their debris–stay where the math tells them to.
Thought experiment: Let's say you are simulating ten thousand satellites on your computer, and the simulation runs until there is a crash. Now let's say the simulation runs for an hour normally. If you increase the speed of the simulation, you get to a crash in a shorter time. Satellites move about 30x the speed of airliners. Hence, if everything else was similar, one would expect 30x the amount of collisions.
> Satellites move about 30x the speed of airliners. Hence, if everything else was similar, one would expect 30x the amount of collisions
Not how orbital mechanics work.
Planes maneuvers, get tossed around and have hubs they circle. A plane under my left wing can’t be relied on to continue in a straight line. The satellite can.
Of course. But your comment that speed is not a factor was incorrect.
Starlink and some other constellations absolutely do maneuver if a close conjunction is predicted https://starlink.com/satellite-operators
Orbits are predictable, but they intersect and decay [at different rates] and occasionally get perturbed by space weather. This already needs periodic conjunction avoidance manoeuvres, and whilst orbits are fast satellite manoeuvres are slow, so the notice you need to avoid a conjunction is measured in hours rather than seconds. Can't imagine a scenario in which it would be sustainable for LEO to even approach the density of commercial aviation, except perhaps for a hypothetical where a single entity actually managed all the satellites.
The other underestimated dimension is that satellite manoeuvres use up a finite supply of expensively-launched propellant. That's tolerable when Starlink is doing 50k conjunction avoidance manoeuvres in six months across its constellation, but once it becomes each satellite moving at least weekly you either need bigger satellites carrying more propellant or have to accept significantly higher collision risk than they currently do.
> and whilst orbits are fast satellite manoeuvres are slow
This is something people unfamiliar tend to misconceive in their limited thinking on the subject. You can't just tap the breaks to slow down. Changing altitude of satellites is done by speeding up to increase altitude and slowing down to lower altitude. Once you change the velocity and reach the desired altitude, you have to then undo that acceleration to get back to orbital velocity. Fuel is required in both directions. The less fuel used the better for the maneuver. Most satellites EoL is defined by remaining maneuvering fuel vs functionality of the hardware.
My first understanding of accelerating in space was from the old Asteroids game. To slow down, you had to rotate 180° and start accelerating in that direction. Others might learn it from Kerbal.
> This is something people unfamiliar tend to misconceive in their limited thinking on the subject
I have a background in astronautical engineering. While you can't tap the brakes to 'slow down', you can impart miniscule amounts of impulse which, over the course of hundreds of orbits, will change your plane by an imperceptible amount from a distance, but tens or hundreds of kilometers up close. OM being OM, you can predicts these collisions in advance.
I had a professor who referred to orbits not in altitude but in expected decay time. We're currently in the months to single-digit years orbits. (We will stay there for telecommunications due to latency.) If we were doing at decades or centuries what we're doing in LEO, this would be a problem. At LEO, it's a nuisance and barely more.
> you can impart miniscule amounts of impulse which, over the course of hundreds of orbits
right. this is what is counter-intuitive for those that are not familiar with space. they don't just light the burner and boost to a new altitude. the part about stopping the acceleration with an opposite burn is often not considered. most think you can fly a space ship like a jet fighter, but in space. can't blame them since that's how sci-fi portrays it. real life space flight is really boring in comparison. jumping out of FTL to land in orbit around a planet makes me laugh every. single. time.
> whilst orbits are fast satellite manoeuvres are slow, so the notice you need to avoid a conjunction is measured in hours rather than seconds
I'm not arguing against collisions becoming more likely. I'm arguing aginst it becoming commonplace to the point that it becomes a commercial concern.
> satellite manoeuvres use up a finite supply of expensively-launched propellant
Nobody is plane changing out of a collision. And for the foreseeable future, in LEO, the birds are not propellant constrained. (And launch is getting cheaper.)
> you either need bigger satellites carrying more propellant or have to accept significantly higher collision risk than they currently do
We're decades away from this being a problem. That gives ample runtime to developing e.g. magnetic station-keeping (if we go reactionless) or more-efficient engines.
> e.g. magnetic station-keeping
I've not kept up for decades now .. what's the state of solar powered magnetorquers these days? I'd quietly assumed it would be more commonplace.
I dimly recall a couple of small satellites magnetically locking fifteen or so years past?
> what's the state of solar powered magnetorquers these days?
Academic. We don't currently have a pressing need for reactionless thrust in the magnetosphere. Each of semiconductors, launch vehicles and telecommunications standards are moving faster than satellites last.
Thanks for the reply.
> Each of semiconductors, launch vehicles and telecommunications standards are moving faster than satellites last.
That's certainly a pragmatic cost based argument for not using them in the fast moving world of commercial magnetosphere constellations.
> Academic.
I feel they've moved past academic and transitioned to deployed .. at some evolution of implementation. Not commercially relevant is certainly one state of play.
I guess I was more interested in the nonlinear control issue in a field of highly variable intensity.
> I'm not arguing against collisions becoming more likely. I'm arguing against it becoming commonplace to the point that it becomes a commercial concern.
Minimising collision risk already is a commercial concern, and the number of conjunction avoidance manoeuvres SpaceX takes in order to achieve this has been growing exponentially (which presumably is a major factor driving their move of 4k satellites to a lower orbit which involves more station keeping) Obviously this gets harder when most of the satellites avoiding their orbits coming too close don't have the same owner, particularly if some of the other megaconstellations aren't even particularly cooperative (hi China!)
> Nobody is plane changing out of a collision. And for the foreseeable future, in LEO, the birds are not propellant constrained. (And launch is getting cheaper.)
No which is why I mentioned the fact that constellations pre-emptively plane change to avoid conjunctions. The frequency with which they have to do this scales superlinearly with the number of satellites operating in or intersecting the orbital plane. Ultimately propellant use for those manoeuvres and station keeping defines the satellite lifetime: agree it's not a huge problem when a satellite is only making small orbital changes a handful of times a year and its got a decent sized delta-v budget for station keeping and EoL deorbiting anyway, but another 70k satellites in the same plane would require quite a lot more adjustments, never mind them operating at aircraft density as proposed earlier.
> We're decades away from this being a problem. That gives ample runtime to developing e.g. magnetic station-keeping (if we go reactionless) or more-efficient engines.
Depends how fast the satellites get put up there (and also whether orbital megastructures become a reality, although non-trivial numbers of them actually might be decades away). There's some scope to improve propulsive efficiency (hi colleagues!), but within the power/mass constraints of a smallsat, you're not likely to see orders of magnitude more improvement in specific impulse over current gen EP, and we are forecast to need orders of magnitude more avoidance manoeuvres, which is generally going to mean more reaction mass. Sure, if we get reactionless propulsion suited for precise orbital changes in LEO then we can forget all about the tyranny of the rocket equation, but hey, if we perfect flying cars we won't have to think about the implications of congestion on the roads!
Make the US land area ~20% larger.
Randomly place 50,000 shoe boxes up and down the entire eastern seaboard.
Randomly place 50,000 shoe boxes up and down the entire western seaboard.
Send them in straight lines towards the other side of the country.
See if any collide. Almost certainly none of them will. Edit: They will almost certainly
For reference, if you placed all 50k boxes next to each other on the same beach, it would be about 10 miles wide. The total shoreline on either side would be ~1800 miles wide.
And that's only 2D.
By my calculations there will be an average of 500 collisions, no? Each shoebox has an effective width of 2 feet, and with 50k of them that's about 1% density. With 50k in the other direction, and about a 1% collision rate, that's 500 collisions.
Seems like you're right, I didn't actually run through the statistics and just went with intuition. Yikes
Yeah my intuition was the same; if I had 1 second to make a guess I probably would have said 1% chance of >0 collisions
This is funnily oddly specific.
Spoken like someone who's never placed a shoebox on a seaboard
If they put their sats low enough (like Starlink already mostly does) any collision debris should be quickly deorbitted by drag, before a cascade can happen.
> cascading collisions
AFAIK they're in separate shells so the probability of collision is basically zero.
Things you put in orbit at a certain elevation don't stay at that elevation forever.
Presumably, there would be a corridor for traveling through elevations whether that was for reaching orbit or de-orbiting. The people placing things in orbit are not doing this with out coordination.
There are in fact many objects that deorbit in an exceedingly uncoordinated manner. It's a statistical inevitability that kessler syndrome is in our very near future if we allow higher orbits to be polluted.
Even if these sats do collide and produce debris, they are in decaying orbits, so that stuff will eventually fall back down.
Debris moves in 3D. Debris moving up will continue moving up. There is no force acting on it to bring it back down. Your comment makes it sound like an explosion would only be in 2D along the same orbit as the original object.
That is not how orbital mechanics work.
It may seem counterintuitive, but if something in orbit gets a push that isn’t strong enough to make it totally escape orbit, it will stay in a new elliptical orbit. That new orbit will pass through the point where the push happened, so it will come back through that location again, just with a different speed and direction.
> There is no force acting on it to bring it back down.
Gravity?
But also orbital dynamics (at least as I understand it) means debris that debris that is flung up is going to have a more oval orbit, so the high point (apogee) increases and the low point (perigee) decreases. And a lower perigee means more atmospheric drag, which will help deorbit the debris.
>means debris that debris that is flung up is going to have a more oval orbit, so the high point (apogee) increases and the low point (perigee) decreases. And a lower perigee means more atmospheric drag, which will help deorbit the debris.
Not quite.
If you are at apogee and accelerate, your perigee will be raised. If you are at perigee and accelerate, your apogee will be raised. You can't increase your apogee and perigee at the same time.
If the impulse is in the direction of orbit, then the altitude of your orbit 180 degrees from your current position will raise. If the impulse is against your orbital direction, your height 180 deg away will be lowered. Once you complete an entire orbit (360 degrees) you will pass through your current position again.
If you wish to move to a higher, circular orbit two impulses are required, 180 deg apart.
That'd have to be one slow explosion to give it less than 1G of acceleration.
According to https://www.youtube.com/watch?v=b66ZZ05wKC0 this might end very badly very soon.
> According to https://www.youtube.com/watch?v=b66ZZ05wKC0 this might end very badly very soon
The paper they cite [1] estimates "the no-manoeuvre collision time" for various orbits. It has no alarming results.
That paper cites another paper [2], which raises the possibility of runaway conditions. It, in turn, runs a model developed in this paper [3].
[1] https://arxiv.org/pdf/2512.09643
[2] https://conference.sdo.esoc.esa.int/proceedings/sdc9/paper/3...
[3] https://www.researchgate.net/publication/234449150_Critical_...
What your describing is called Kessler Syndrome
https://en.wikipedia.org/wiki/Kessler_syndrome
... It is a very real possibility, but less of a problem below 550km altitude because the decay time is much shorter (and why all of these mega constellations tend to stay at lower altitude, even though ~1000km is generally better for a communications satellite).
> It is a very real possibility
It's really not. Not in the popularly-portrayed manner. Militaries have been researching how to intentionally cause such a cascade in even a limited orbit. To my knowledge, there isn't a solution.
It’s like announcing you’re going to sell corn starting a year from now when your competition owns all the land corn is grown on, started selling their corn 6 years ago, and has gotten really good at making it cheaper and producing more, based on real world experience.
And you’ve only used a super complex corn planting robot twice, and you need to do it thousands of times
Might be better to replace url with the full press release which has actual information
https://www.blueorigin.com/news/blue-origin-introduces-teraw...
>The TeraWave architecture consists of 5,408 optically interconnected satellites in low Earth orbit (LEO) and medium Earth orbit (MEO).
An even better source, imho:
https://arstechnica.com/space/2026/01/blue-origin-we-want-to... ("Another Jeff Bezos company has announced plans to develop a megaconstellation")
I'll put that one in the toptext as well - thanks!
OK, we've used that link instead and put the submitted URL (https://www.blueorigin.com/terawave) in the toptext. Thanks!
From a technical standpoint: amazing achievement, and the tech nerd in me is in awe. But it feels like a lot of people don't understand (or care?) how much these companies are polluting the space.
Before the "new wave", in 2010-2015 or so, Earth had around 1500 active satellites in orbit, and another 2,000-2,500 defunct ones.
Starlink now has almost 9,500 satellites in orbit, has approvals for 12,000 and long-term plans for up to 42,000. Blue Origin has added 5,500 to that. Amazon plans for 3,000. China has two megaconstellations under construction, for a total of 26,000, and has filed for even larger systems, up to 200,000 satellites.
We might be the last generation that is able to watch the stars.
> We might be the last generation that is able to watch the stars.
I'm not convinced this is a major issue, but I'd like to hear arguments for why it is.
Correct me if I'm wrong, but aren't LEO satellites only going to reflect light from the sun when they're at low angles near sunrise and sunset? For night time stargazing, they're going to be in Earth's shadow, too.
The amount of light they reflect back is also small. They can be seen if you look closely at just the right time, but I don't understand how this is supposed to be so much light that it starts raising the overall background light level considerably. The satellites are small and can only reflect so much.
Is it just annoyance that they're up there and showing up in photos?
> Correct me if I'm wrong, but aren't LEO satellites only going to reflect light from the sun when they're at low angles near sunrise and sunset? For night time stargazing, they're going to be in Earth's shadow, too.
Iridium's LEO satellites were sometimes (impressively) visible after midnight.
"Polluting" is a very charged term. These satellites provide immense value. So far, there is no evidence these will stop us from watching the stars.
"value" is also a very charged term, and pollution is nearly purely a byproduct of the pursuit of value.
(Also, for a frame of reference as to how large these numbers are: the entire gps network operates on 31 satellites.)
Is it a lot? It's a bit like you are telling me there are gonna be 250000 cars on a planet larger than Earth.
With the difference that cars can steer and stop to avoid collisions and aren't necessarily in your field of view every time you look at the night sky ;)
I have no idea if the number is actually a lot shrug but it's surely different than cars on a planet's surface
LEO Satellites are only visible after dawn and before sunrise. They are invisible to the eye and even large telescopes when they are not in sunlight.
I wonder if there's a limit to space junk beyond which leaving the Earth in a space shuttle becomes impossible.
These satellites are low Earth orbit (LEO)
They're extremely sparse. Imagine putting 12,000 satellites randomly over the surface of the Earth. You're just not going to bump into one, statistically. Now expand that into 3D space in an orbital zone above us.
It's not a collision risk.
It is already impossible - all the remaining Space Shuttles are in a museum, not to mention all Space Shuttle missions were (and were always intended to be) to Earth orbit. No Space Shuttle ever went past 600 km hight Earth orbit.
> wonder if there's a limit to space junk beyond which leaving the Earth in a space shuttle becomes impossible
There is. We don't have the industrial capacity, as a species, to do it.
Not to mention low orbit being self cleaning and higher orbits being exponentially more space. You can map the junk with radar & plot the launch to avoid it.
How many causes Kessler syndrome?
> How many causes Kessler syndrome?
Space is huge. Try this trick: the number of satellites in orbit is about the same as the number of planes in the air at any time. (~12,000 [1].)
The volume of space from the ground to 50,000 feet is about 200x smaller than the volume from the Karman line to the top of LEO alone (~2,000 km).
Put another way, we approach the density of planes in the sky in LEO when there are milliions of satellites in that space alone. Picture what happens if every plane in the sky fell to the ground. Now understand that the same thing happening in LEO, while it occurs at higher energy, also occurs in less-occupied space and will eventually (mostly) burn up in the atmosphere.
Put another way, you could poof every Starlink simultaneously and while it would be tremendously annoying, most satellites orbiting lower would be able to get out of the way, those that couldn't wouldn't cause much more damage, the whole mess would be avoidable for most and entirely gone within a few years.
There are serious problems with space pollution. Catastrophic Kessler cascades that block humans from space, or knock out all of our satellites, aren't one of them.
[1] https://www.travelandleisure.com/airlines-airports/number-of...
You're ignoring the speed they're travelling at.
For a given period of time, a single satellite will travel through a vastly larger volume of space than a single plane.
> You're ignoring the speed
Nope.
> a single satellite will travel through a vastly larger volume of space than a single plane
Linearly (with a cap). Volume grows faster.
At the altitudes these mega-constellations operate at, kessler syndrome is not a real threat. Even if left unpowered, everything there will naturally re-enter the atmosphere in ~5 years.
5 years of no Starlink is a serious threat, especially to SpaceX.
Looks like they are using lasers for backhaul down to ground stations. What happens if the beam is obstructed for a brief moment (plane, kite, ufo, etc..)?
> What happens if the beam is obstructed for a brief moment (plane, kite, ufo, etc..)?
Same as with any dropped packet.
And my guess would be multiple beams for redundancy.
All those AI datacenters in space will need a way to get data to them.
Bezos can't even build his first constellation and already planning his second... Possibly the real play here is snapping up more frequency licenses on earth (we need them because we're launching any day now promise). They are the real constraining resource and could be used to keep others out of the market for a while.
> They are the real constraining resource and could be used to keep others out of the market for a while
I'd love to see a betting market on a unified, global licensing regime lasting for another ten years.
Well, this shoots down the argument that Blue Origin is just Amazon's space wing. Strange to see them launching a direct Amazon Leo competitor but now that they have reusable boosters (on paper) it does make more sense for them to control the lion's share of their launch manifest with their own megaconstellation.
It's not a Leo competitor, it's a different type of service offering.
Fair enough, I was misremembering the higher end of Leo's bandwidth (1Gbps down/400Kbps up). Doesn't really intersect with TeraWave at all
So the difference here is bandwidth throughput/speed?
It's a dual modem solution, optionally. Laser or RF. Different speeds and reliability levels.
Target customers of the Government, Enterprise, etc type.
Latency may play a factor here, I'm not sure at which height they plan to put them.
this seems rather expensive but i get that its not competing with spacex here for consumer market