I've seen people saying that things like Starlink are inherently safe because they use very low orbits.
But higher orbits are just higher speeds and when two objects collide and break apart some smaller fragments might fly off with speeds far greater than those of colliding objects.
So failure at very low orbit might add a lot of debris to higher orbits that might trigger Kessler syndrome.
> But higher orbits are just higher speeds and when two objects collide and break apart some smaller fragments might fly off with speeds far greater than those of colliding objects.
Objects in higher orbits are slower[1] than objects at lower altitudes, and debris from a hypothetical collision between two high-orbiting objects would not accelerate to a speed "far greater" than the originating objects, because that increase in speed would require some form of acceleration. One satellite smashing into another would not add sufficient energy to the debris to alter its speed significantly. Any change to the speed of the resulting debris, if it were to occur, would also necessarily alter its orbit. The object's speed and its orbital altitude depend on each other.
> So failure at very low orbit might add a lot of debris to higher orbits that might trigger Kessler syndrome.
It should be noted that "failure" does not mean "exploding into a thousand pieces", so a malfunctioning satellite by itself is not that much of a problem. Low altitude satellites such as Starlink are actually better from a space junk perspective — their altitude means the effect of atmospheric drag on the spacecraft will eventually (within months, sometimes) cause it to re-enter the atmosphere instead of just hanging around in Earth orbit forever, like dead satellites do at higher orbits.
Kessler syndrome would be more likely to occur with satellites in medium and high Earth orbit, since there is no drag to bring the dead spacecraft or debris back down to Earth. Anti-satellite weapons are particularly dangerous here, because they're likely to target high-orbiting satellites and leave debris fields behind.
---
[1] As pointed below, in terms of pure kinetic energy this is not correct. I was referring to apparent speed vs an object in low-Earth orbit.
> Objects in higher orbits are slower than objects at lower altitudes
You are right. My point still stands. On collision debris might have significantly different velocities than the original objects thus have different orbits.
> One satellite smashing into another would not add sufficient energy to the debris to alter its speed significantly.
I think that's completely dependant on the mass of the specific particle of debris. The smaller the particle, the larger possible speed change attainable in a messy collision.
I'd really prefer that there was some scientific study simulating possible results of impact between two SpaceX satellites. Because there's non-zero chance of that happening. For example if there are some errors during the deployment of new batch.
Yes, sorry, my phrasing is somewhat confusing. In terms of energy they are faster, since more speed is required for higher orbits. They appear to be slower in a relative sense when observed from Earth — the Space Station or a Starlink sat will appear to be moving significantly faster than a satellite in medium or high-Earth orbit.
It also depends on the geometry of the orbit — an object in a circular orbit will move at a constant speed, while an object in an elliptical orbit will move faster or slower depending on which point of the ellipse it is.
I understood the parent comment to be focused on an Earth-centred perspective, since they were talking about Kessler syndrome and Earth-orbiting debris. I was trying to address that point from an Earth-centred perspective too, but I appreciate how that can be confusing.
No, no. You were right about the speeds and I was wrong. Higher circular orbits do have lower speeds. Starlink satellites have speeds over 7km/s while satellites on much higher geostationary orbit have little over 3km/s. I haven't built my intuition about why it's true but it is. Orbital speed is proportional to inverse square root of orbit radius.
One thing you may not have considered is the shape of the orbit of the debris. While the orbit of the aggregated debris pre-disassembly was likely mostly circular, debris that heads upward (or downward) post-disassembly will end up in an eliptic orbit, which will mean it spends some time above the previous orbit and some time below.
While the time above raises risk of impacting objects in higher orbit, the time below has increased atmospheric drag and reduces the time in orbit. The higher the apogee, the lower the perogee, but I haven't looked at detailed analysis.
Those satellites orbits are roughly on the exactly the same sphere, so if two of them collide at an angle debris would shoot out tangentially to this sphere so while apogee would be higher, perogee would be at the same distance as original orbit.
Kuiper and Blue Origin have been pretty entertaining. Bezos can throw all the money in the world at space but it still hasn't bought him a single functioning satellite.
Their original plan for Kuiper was and seemingly still is to launch the first two satellites in Q4 2022, so it's too early to declare them failed. (Their orbital rocket, New Glenn, however, has been delayed a few times.)
No, their first orbital class rocket is New Glen, and I think the consensus is they are still a couple of years away from launch.
The engine for New Glen, BE-4, will also be used in the new ULA Vulcan rocket. It is expected to launch in the next year or so too but is running at least three years behind schedule. There is a chance that engine will reach orbit on Vulcan before New Glen launches at all.
By the time Amazon is relevant Starlink will have already cornered the market. Not to mention keeping such a constellation maintained requires crazy launch cadence which takes atleast a decade to hone even if your rocket is really damn good out the gate, not to mention if your rocket isn't fully reusable you probably just won't be able to afford to properly maintain a mega-constellation.
i.e I'll believe it when I see it. For now the only thing Bezos space companies have been producing is hot air.
The number of concurrent users is limited by the operator’s ability to reuse frequencies across geographies, which in turn is limited by the angular resolution of satellite antennas and satellite altitude (ceteris paribus).
Inmarsat is mostly marketing to ships at sea, and other cases where users are quite far from each other. So they can use a small number of geostationary satellites.
Starlink is marketing to “regular people” in exurban and rural places that are uneconomic to serve with DSL or cable. So they need to create very small cells, hence low altitude satellites. But low altitude satellites can’t see as much of the ground at once, so they need a lot of them.
Also, the geostationary model doesn’t provide good coverage of high latitudes (and launching geosynchronous satellites in high inclination orbits to cover those high latitudes would multiply the number of satellites needed by Inmarsat). As a result, there are basically two models for providing satellite comms: a few geostationary or a bunch of polar LEO.
The title of this post is not the title of the article, the article doesn’t even mention Space X. There is “editorialising” (which HN discourages) and then there is just sticking a clickbait title on a post to get a debate going.
The thing is, the topic of the article, Amazon’s Project Kuiper winning a DoD contract is interesting to many reasons. It’s shows the DoD ensuring there are other operators other than SpaceX, but also the dedicated “mesh network” for connecting military assets.
It’s almost an interesting experiment, at the time of posting none of the other 9 comments mention the content of the article.
Current title: Watch out Starlink, Amazon is coming for you
Correct title: Amazon to link Kuiper satellites to DoD’s mesh network in space
> Amazon’s Project Kuiper will install DoD-compliant laser communications terminals on its internet satellites so they can transfer data from remote-sensing satellites directly into the military’s mesh network in low Earth orbit.
I find this deeply concerning that the military is taking more and more dependencies on Amazon, and would therefore have an interest in Amazon’s survival.
I spent almost 11 years at Amazon, and exited after I couldn’t bear with my mental and physical health deteriorating any more.
It’s a toxic, abusive culture. Their real innovation is how to manipulate and abuse employees, at scale, without destroying your business reputation.
What we ought to do is break up Amazon, but instead, we’re rewarding its culture of extreme narcissists and sociopaths.
18 comments
[ 2.7 ms ] story [ 44.9 ms ] threadBut higher orbits are just higher speeds and when two objects collide and break apart some smaller fragments might fly off with speeds far greater than those of colliding objects.
So failure at very low orbit might add a lot of debris to higher orbits that might trigger Kessler syndrome.
Am I wrong about this?
Objects in higher orbits are slower[1] than objects at lower altitudes, and debris from a hypothetical collision between two high-orbiting objects would not accelerate to a speed "far greater" than the originating objects, because that increase in speed would require some form of acceleration. One satellite smashing into another would not add sufficient energy to the debris to alter its speed significantly. Any change to the speed of the resulting debris, if it were to occur, would also necessarily alter its orbit. The object's speed and its orbital altitude depend on each other.
> So failure at very low orbit might add a lot of debris to higher orbits that might trigger Kessler syndrome.
It should be noted that "failure" does not mean "exploding into a thousand pieces", so a malfunctioning satellite by itself is not that much of a problem. Low altitude satellites such as Starlink are actually better from a space junk perspective — their altitude means the effect of atmospheric drag on the spacecraft will eventually (within months, sometimes) cause it to re-enter the atmosphere instead of just hanging around in Earth orbit forever, like dead satellites do at higher orbits.
Kessler syndrome would be more likely to occur with satellites in medium and high Earth orbit, since there is no drag to bring the dead spacecraft or debris back down to Earth. Anti-satellite weapons are particularly dangerous here, because they're likely to target high-orbiting satellites and leave debris fields behind.
---
[1] As pointed below, in terms of pure kinetic energy this is not correct. I was referring to apparent speed vs an object in low-Earth orbit.
You are right. My point still stands. On collision debris might have significantly different velocities than the original objects thus have different orbits.
> One satellite smashing into another would not add sufficient energy to the debris to alter its speed significantly.
I think that's completely dependant on the mass of the specific particle of debris. The smaller the particle, the larger possible speed change attainable in a messy collision.
I'd really prefer that there was some scientific study simulating possible results of impact between two SpaceX satellites. Because there's non-zero chance of that happening. For example if there are some errors during the deployment of new batch.
It also depends on the geometry of the orbit — an object in a circular orbit will move at a constant speed, while an object in an elliptical orbit will move faster or slower depending on which point of the ellipse it is.
I understood the parent comment to be focused on an Earth-centred perspective, since they were talking about Kessler syndrome and Earth-orbiting debris. I was trying to address that point from an Earth-centred perspective too, but I appreciate how that can be confusing.
While the time above raises risk of impacting objects in higher orbit, the time below has increased atmospheric drag and reduces the time in orbit. The higher the apogee, the lower the perogee, but I haven't looked at detailed analysis.
So the debris does not build up over time.
< 500km -> 25 years 800km -> 100-150 years 1200km -> 2000 years
https://www.aboutamazon.com/news/innovation-at-amazon/projec...
The engine for New Glen, BE-4, will also be used in the new ULA Vulcan rocket. It is expected to launch in the next year or so too but is running at least three years behind schedule. There is a chance that engine will reach orbit on Vulcan before New Glen launches at all.
i.e I'll believe it when I see it. For now the only thing Bezos space companies have been producing is hot air.
https://gcaptain.com/inmarsats-new-global-broadband-network-...
Inmarsat is mostly marketing to ships at sea, and other cases where users are quite far from each other. So they can use a small number of geostationary satellites.
Starlink is marketing to “regular people” in exurban and rural places that are uneconomic to serve with DSL or cable. So they need to create very small cells, hence low altitude satellites. But low altitude satellites can’t see as much of the ground at once, so they need a lot of them.
Also, the geostationary model doesn’t provide good coverage of high latitudes (and launching geosynchronous satellites in high inclination orbits to cover those high latitudes would multiply the number of satellites needed by Inmarsat). As a result, there are basically two models for providing satellite comms: a few geostationary or a bunch of polar LEO.
The thing is, the topic of the article, Amazon’s Project Kuiper winning a DoD contract is interesting to many reasons. It’s shows the DoD ensuring there are other operators other than SpaceX, but also the dedicated “mesh network” for connecting military assets.
It’s almost an interesting experiment, at the time of posting none of the other 9 comments mention the content of the article.
Current title: Watch out Starlink, Amazon is coming for you
Correct title: Amazon to link Kuiper satellites to DoD’s mesh network in space
I find this deeply concerning that the military is taking more and more dependencies on Amazon, and would therefore have an interest in Amazon’s survival.
I spent almost 11 years at Amazon, and exited after I couldn’t bear with my mental and physical health deteriorating any more.
It’s a toxic, abusive culture. Their real innovation is how to manipulate and abuse employees, at scale, without destroying your business reputation.
What we ought to do is break up Amazon, but instead, we’re rewarding its culture of extreme narcissists and sociopaths.