Yeah, that fine should be two or three orders of magnitude higher, but it’s a good start that they’re doing it at all.
We really do need stronger requirements for satellites. One good thing the FCC did was require that satellites under 2000 km must deorbit in <5 years. But even that is probably not enough when we have 100,000 satellites in orbit. Satellites should be obligated to have onboard propulsion over a certain size. And the FCC should probably have requirements for both average and minimum deorbit time. Make the average requirement smaller yet–say, a year. 5 years is fine as the maximum, as long as the vast majority deorbit faster than that.
As always, there’s de-orbit if it’s still fully capable under active ground control vs de-orbit if it becomes computer-bricked or mechanically crippled.
Requiring prompt active de-orbit capability within a year for a fully functional satellite seems like a good idea.
Requiring some kind of very robust dead-man passive device that could still function if the comm is down long-term, the on-board computers are dead, or the propellant has leaked away would be another good idea. I’m thinking a spring-deployed drag tether that is continuously prevented from deploying by a heartbeat from the primary computer system. And which is compact enough that even after a collision, that component at least would probably be intact enough to de-orbit itself and whatever wreckage it’s still attached to.
Agreed that something like that could be a good idea, though I wouldn’t want the FCC (or whoever) to specify the details too closely.
With Starlink, it works like this:
If the propulsion system is operational, it actively deorbits in months
If the propulsion system is out, but the gyros are working, then it orients itself in a high-drag configuration (with the solar panels acting as a sail) and deorbits in about a year (as I recall)
If everything is out, the sats still have a pretty high average ballistic coefficient, and it’ll come down in 5 years
The average obviously depends on the reliability of the components, but assuming reasonable values (i.e. total failure is not the common case), it should average under a year.
But some of that is specific to Starlink, so tethers and similar things could prove useful as well.
In fact, the first cubesat I worked on had just that sort of thing, although it was a balloon, not a tether. A large mylar balloon inflated with a small CO2 cartridge, and the software had just that kind of dead-man functionality that would trigger it if it didn’t receive comms for a while. It didn’t work out for various reasons, but the principle was sound.
So the maximum lifespan of any satellite is to be 1-5 years, with the low end dominating? I don’t think that will work. Satellites today have an average lifespan of 15 years. There is some talk of lowering that to 7-8 years, but much lower than that and many of them no longer have a viable financial model.
It costs a lot of money to design, build, test, fly, operate, and decommission a satellite. The chance that they would be able to recoup their costs in a year or two or three seems unlikely.
When SpaceX’s Starlink constellation is finished, It will have somewhere in the neighborhood of 40,000 satellites in space. There is no way those could be replaced every year or two. Even at 5 years of lifespan, that’s 12,000 satellites put in orbit per year to maintain the system. It’s not going to get shorter than that.
A couple other things that need to happen as well:
A one-stop-shop for monitoring active satellites and notifying operators that they need to do something. Ideally, totally automated, and with rules for who has the right-of-way. Possibly with a bidding process. Each satellite in a conjunction could respond–but it might be cheaper for one of them to do so. The other satellite can pay them their share of the maneuvering costs.
Debris removal vehicles for everything that slips through the cracks. Not sure who pays for this. Mandatory insurance for satellite operators? Actually, much of the problem is with the launch providers and the upper stages they leave in orbit, not just the satellites.
safety deposit of 5x the cost of rescuing a satellite to be deposited to agency XYZ before launch … and agency XYZ will take care of your potential problems…
money to be returned at the end of the lifecycle of the satellite (excl. satellites with incidents)
Optimally, I would like to see recycling and manufacturing operations in orbit that make it cost effective to run a drone fleet collecting leftovers and other debris for recycling. That’s a ways off, but I think it should be a priority. Anything up there that can be reused or recycled is valuable; any usable material is worth over $2200/kg, just in terms of launch cost. The Falcon Heavy second stage, for example, has an empty mass of roughly 4000kg, so each is potentially worth $8.8 million, minus the cost of moving it from its post-separation orbit to an orbital depot. (The second stage is typically left in a rapidly decaying orbit, so it’s not part of the long-term space junk problem, but it wouldn’t have to be if there were an orbital facility that could use the material.)
Retrieving the junk and moving it between orbits is not trivial, of course, but the fact that it doesn’t have to be done quickly should simplify the problem. I picture a fleet of small drone tugs with EM tethers and solar panels doing most of the work, to minimize use of reaction mass, but there may be better ways.
The big junk is dangerous of course, but there is relatively little of it.
The quantity of stuff in orbit goes up inversely with size according to a power law. So there may be 1,000 pieces of debris around a meter across, 10,000that are 10cm or so, 100,000 that are 1cm, and so on. By the time you get down to the size of paint flakes or sand particles, we are talking many billions or even trillions of objects. And even a paint chip going orbital velocity can do significant damage.
Clearing orbital space is a hellaciously difficult problem. We are talking about a volume with a surface area bigger than the Earth, but thousands of km thick. And in it we are trying to catch stuff that carries enormous energy, and in the dase of the really small stuff, is untrackable.
One potential future mitigation is making non-frangible satellites so they don’t blow into a zillion pieces when they collide with something. What those would look like, I’m not sure. But the danger isn’t really with the big satellites - there are few enough of them that they rarely hit something, and we track them so we can usually avoid being near them. It’s the millions of pieces of debris from satellites that do collide that are the real threat over time.
Here’s a good reference for orbital debris. although all the amounts are bigger now than when this was published in 1995:
The first two test satellites for Amazon’s Kuiper satellite internet system out of over 3200 satellites in total–a competitor to Starlink.
Frankly I never really understood why Amazon is in this business at all–will satellite end up with more than a couple percent or so of the internet delivery business compare to terrestrial wired and wireless systems?
The total global ISP market is about $1.3T in revenue. So even a few percent isn’t chump change.
Fast satellite internet is also preferentially useful for high-dollar applications like military, airlines, ships, etc. They can capture a much higher fraction (100% in some cases) of those markets.
It’ll be interesting to see if the new availability of fast internet in rural locations drives population movement. Expensive housing is a huge problem in some areas… but if people can work from home in cheap rural areas because they can now get decent internet there, that may drive even more people there, which expands the market for satellite internet even further.
Amazon will have their work cut out for them. For some reason they’ve refused to launch on SpaceX rockets, and signed contracts with three rockets that haven’t even launched once yet. They have a few launches signed for the Atlas V, which is at least a functioning rocket, if expensive. Amazon’s shareholders are suing them for not considering SpaceX for launch. For their part, SpaceX has happily launched for most of their competitors so far: OneWeb, ViaSat, Iridium, and others.
I’ve been looking for real estate for exactly that reason. A small place either off-grid or at least outside of normal internet coverage, which can be expected to go up in value once Starlink is widely available. A place with Starlink and a nice view could also make a good Airbnb.
Unfortunately, prices in most places are still elevated due to the pandemic.
Changing inclination will be the big problem. But yeah, perhaps there are propellant-free techniques can get stuff lined up nicely enough to be useful. Some of the debris itself can be used as propellant, like maybe an ion thruster using fine aluminum particles.
I agree that inclination changes are likely to be the biggest hurdle (aside from locating the junk to be collected in the first place). They’re expensive in terms of delta-v. My hope is that the fact that the orbital adjustments don’t need to be made quickly will open the way for gradual, propellant-free adjustments. Perhaps it would be possible to leverage the drones’ solar panels as sails to nudge them into gradual spirals toward the desired inclination? With the EM tether handling altitude maintenance, light pressure and solar wind on the panels should be the primary remaining forces acting on the drone. With only the limited surface area of the panels, it would be very slow, but it might be worth it.
Cutting propellant use to the absolute minimum would make re-use of waste material as propellant for critical maneuvers more viable, since you wouldn’t need as much of it. Maybe have two sets of drones–the bulk of the fleet could be tether-sail tugs, with a smaller set that augment their maneuvers with ion thrusters that would be deployed in a targeted fashion for junk the smaller drones spot, but can’t manage efficiently.
(I realize that this may be a weird thing for me to be excited/obsessive about, but I really think that we need recycling and manufacturing facilities in space to provide a solid foundation for large-scale exploration and development.)
what if you pull that thru, b/c starlink, etc… and then once starlink has a couple of 100.000 users captivated increases their price by the factor of 5 or 10 … or worse: let’s stop it altogether, b/c jews …
from a risk analysis pov … are you really comfy with your project hinging on the mental state of loony Mr. M?
Or of them simply underprovisioning versus demand? In the early days, the requirement to offer 24h services demands they oversupply satellites and bandwidth versus their tiny customer base. Once that base is not tiny, the economics run the other way: the less they provide the higher the profit as long as folks aren’t quitting. But with no competitor, they can’t quit; instead they’ll just suck it up.
And especially for folks near the latitude extremes, the dwell time up there and satellites visible will be fewer than in more equatorial places. I don’t know what the advertised latitude limits are for the StarLink constellation, but I would not want to be near them and also near any concentration, even a small one, of population.