"Biggest advantage of lower altitude is that beam diameter is smaller for a given antenna size."
See full article...
See full article...
SpaceX begins “significant reconfiguration” of Starlink satellite constellation
- Thread starter JournalBot
- Start date
I wonder if this means they don't expect to have Starship up to operational cadence anytime soon? Because IIRC the original plan was to have much larger satellites fly on Starship to give smaller spot beam size.
Upvote
39
(43
/
-4)
14,000 active satellites of which, 8000 are Starlink satellites. Is that right, more than 57% of active satellites are Starlink?
How many inactive satellites are there?
How many inactive satellites are there?
Upvote
107
(107
/
0)
I think the primary reason is for ensuring quicker deorbit times in case of a satellite failure. The "performance" improvement was highlighted by Musk - so it sounds like he's trying to gild the lily by saying that it will also improve latency as a side effect.
Upvote
70
(75
/
-5)
These satellites should be EOL by then. I think they're only planned to be up there for five years.
Upvote
102
(102
/
0)
Although packed more tightly, the Starlink satellites won't be packed that much more tightly. The satellites are moving from a 341 mile to a 298 mile altitude; the radius of the Earth is about 3960 miles, so orbital radius changes from 3960 + 341 to 3960 + 298 miles, or from 4301 to 4258 miles. This is a change of 1%, and thus the resulting spacing between the satellites also changes by 1%.
Upvote
184
(190
/
-6)
Not just latency but speed too. That's because each satellite's spot beam is now smaller and therefore serves fewer customers, making average speed per customer in the beam go up.
Upvote
85
(86
/
-1)
Newbie question: Is Starlink reaching limits of scale?
Ie. A narrower beam > fewer customers per beam (each satellite handles fewer clients/streams at once).
Upvote
27
(28
/
-1)
The spot beams have a fixed 3 degree angle. They will illuminate a smaller cell on the ground once operated at a lower altitude.
Upvote
122
(122
/
0)
https://www.wolframalpha.com/input?i=time+for+speed+of+light+to+travel+43+miles
0.2 milliseconds.
Couple decades back, a friend of mine worked at a NOC for one of the backbone networks. He was given an escalated support ticket from a customer who was complaining about latency between the east and west coast of the US being 15 milliseconds. He looked into the network, then paused, did the math, informed the customer that the speed of light limited latency to no less than 13.9 milliseconds, and closed the ticket.
This customer, I suspect, now uses Starlink and will the thrilled at the 0.2ms improvement.
Upvote
213
(215
/
-2)
The area over which the satellites are distributed varies with the square of the radius, so a 1% change in radius decreases the area by (1.0 - 0.01)^2. Since the change is small, it effectively amounts to a 2% decrease in spacing, not 1%. But your point remains valid, it is a pretty small change.
Upvote
156
(158
/
-2)
Starlink went from 4.5m users at the end of 2024 to over 9m users now. SpaceX is definitely worried about congestion, with many areas already over-saturated with customers and new customers having to pay an extra one-time fee just to connect.
So yes, Starlink is reaching a bit of a limit, and this plan by SpaceX can bring out a bit of extra performance in high-demand areas. However, as the article says, the V3 satellites will massively expand capacity. They just need Starship to be operational first.
Upvote
103
(103
/
0)
Musk said something similar, and I’m missing something. If there’s still the same number of subscribers and the same number of satellites, then won’t each satellite be serving the same number of users?
Think about it the other way: if the satellites were already at a lower attitude, and then the satellites were magically, instantly raised higher, each satellite could serve exactly the same users they could at the lower altitude right? Just because they can theoretically “see” other users doesn’t mean that they have to serve those other users (that’s the whole point of the beam shaping, etc).
Upvote
13
(20
/
-7)
Yeah, and you're potentially losing overlap if multiple satellites were able to provide failover/bandwidth.
Upvote
18
(20
/
-2)
There's an entire RF tower chain from NY to Chicago that was setup so that some high frequency traders could arbitrage the markets a bit faster than folks using fiber optics. I'm not sure if Starlink with RF up and down plus laser links through vacuum is faster than the RF through air plus repeaters, but if it is I'm sure someone paid Musk $$$$$$$ to use it for similar purposes (come to think of it, I'm sure that Starlink could be used to do such work between NY and London where RF relay isn't possible).
Upvote
70
(71
/
-1)
Mungus the Unhyphenated
Ars Tribunus Militum
Large solar flares can occur at any time, so even at the true solar minimum in the sunspot cycle, the risk never really goes away. What we're dealing with here is the change in ionospheric activity, and its knock-on effects on the atmosphere, that does follow along with the sunspot cycle.
It's cool that we're now able to move entire constellations of satellites in response. It wasn't that long ago when satellites had only enough maneuvering ability mainly to hold position, with minimal ability to re-position over their lifetimes.
Upvote
58
(58
/
0)
Quote adjusted to make a point.
There is always some new quality of life improvement that comes along and someone complains about how they didn't need that in their day, or something similar.
Starlink does provide an invaluable service for lots of people around the world. So much of life assumes you have internet these days, those without it at are a major disadvantage.
Upvote
132
(139
/
-7)
There are only about 3000 inactive payloads on orbit, plus a bunch or rocket bodies and other debris. About half of those are Soviet or Russian. And most of them are pretty old.
Most current launch and satellite operators are pretty good about cleaning up after themselves.
Upvote
76
(76
/
0)
Did we really?
Humankind did pretty horrible stuff before then. Also doing pretty horrible stuff since.
Upvote
24
(35
/
-11)
Stipulating that it's probably a dumb question, and acknowledging the slower (but still accelerating) pace of its development, but wouldn't a warming planet create the same kind of effects at altitude? Warmer gas expands, after all.
Since we're talking density instead of turbulence (as in ripping away some of the atmosphere to higher altitudes where satellites orbit from CME's) wouldn't a warming earth create the same issues, even if at a slower pace?
Not trying to troll this. It seems like a hotter atmospherics will expand, and that the solar storms lead to heating in the upper atmosphere which increased the altitude at which the atmosphere can interfere with satellite orbits. If that's the case, I'd think climate change would do the same thing. I don't know by how much or how fast (even if the pace of climate change is increasing).
I also get that the satellites up there in that orbit NOW probably won't be the same ones later if the atmosphere expands due to warming to the altitudes they do during solar maxes. So it's kind of a thought experiment at this point. I am being a bit lazy asking about it, but I think it's a bit of trivia that, if not relevant today, will be sooner than one might expect. So assuming I'm not misreading this situation, does anyone know the math for how soon that might be, or at what global surface temperature it would most likely occur?
Upvote
3
(13
/
-10)
The users aren't evenly distributed across the entire planet and there are limits to the number of sats you can have pointing at a given area. The dishes need to be able to differentiate between the signal coming from Sat A vs SAT B. That could be from them being on different frequencies or being spaced far enough apart for the dish to be able to differentiate between them even on the same frequencies.
The result of the smaller beams will mean more useable ones pointed at an area of congestion. If at their old altitude NY city had 10 beams pointed at them and at the new altitude they have 12 that's more bandwidth for everyone in NYC to share.
My personal take is this is the primary reason for this and the safer is positive side effect that they can spin too. This may end up shortening their useful life but with v3 on it's way soon(tm) they probably don't care. Get more use out of them now and retire them early for something bigger and better.
Upvote
34
(37
/
-3)
NYC would be very high on my list of places where Starlink is a largely unnecessary service.
Upvote
67
(68
/
-1)
The beam size - or Earth footprint - is the crucial factor in allowing Starlink to serve more customers in the more dense regions of the planet, so I suspect that is the priority. Lower latency is serendipitous - a nice-to-have - but I don't downplay the "self-cleaning" aspects of those lower orbits; Starlink's future depends on making sure that defunct spacecraft do not impair operations....or threaten other satellite.
Upvote
32
(32
/
0)
Couple of questions ...
1. I know that ion engines are efficient, but how will the lower orbits affect these satellites' lifetimes? The article mentions that a non-responsive satellite would deorbit within months compared to years at the higher orbit. So the ion drives will need to use more fuel to counteract atmospheric friction.
2. The lower 300-mile altitude brings these satellites closer to the orbital altitudes of the ISS and Tiangong space stations, which orbit between 220-280 miles. Will this increase the debris collision risk to the stations?
1. I know that ion engines are efficient, but how will the lower orbits affect these satellites' lifetimes? The article mentions that a non-responsive satellite would deorbit within months compared to years at the higher orbit. So the ion drives will need to use more fuel to counteract atmospheric friction.
2. The lower 300-mile altitude brings these satellites closer to the orbital altitudes of the ISS and Tiangong space stations, which orbit between 220-280 miles. Will this increase the debris collision risk to the stations?
Upvote
5
(8
/
-3)
Our greenhouse gases and other climate meddling seem to be causing warming of the lower atmosphere and significant cooling of the upper atmosphere.
Upvote
45
(45
/
0)
What you're missing is that at the lower altitude they can use more spot beams.
As the number of beams goes up, so does the total throughput, because each one has a fixed amount of bandwidth.
Upvote
32
(33
/
-1)
Even in cities there can be spots missing good broadband access but it was more of just an illustration of the concept. That said even a single percent of the population in NYC might be more than Starlink can support given the density of people there.
An area with a lot of potential customers and more spot beams in the same area = more bandwidth for those customers.
Upvote
17
(20
/
-3)
Sending a modulated neutrino beam would be even faster, especially for more widely dispersed cities, because you'd get essentially the speed of light along a straight line instead of following the curve of the surface of the Earth. Building a suitable detector is left as an exercise for the reader, but I'm sure one could get some folks to invest in such a scheme.
Upvote
69
(71
/
-2)
The effect of climate change on thermosphere thickness and density is less than a rounding error, and it might be directionally inverted from what you are expecting.
The heat that impacts the thermosphere is absorption of far UV photons from the sun, which have the energy to knock electrons off their atoms, ionizing a significant fraction of the wispy gas particles. You can find pages asserting that the upper thermosphere, around the same altitude as Starlink, has a “temperature” of 2000 C, but that’s an imprecise description because the gas particles aren’t in a true thermal equilibrium where there’s a Maxwell Boltzmann distribution of particle velocities. Compared to a MB distribution, the particles in the thermosphere have a long tail with more ultra fast particles that would exist once the particles thermalized through repeated collisions. The gas is so rarified that particles can go entire seconds between collisions, and UV photons are ripping electrons off their atoms, sending them flying and restocking the population of ultra fast charged particles faster than thermal interactions can restore a thermal equilibrium.
The Sun’s visible light output varies by less than 0.1%, being dominated by thermal emissions from the sun’s visible surface (photosphere), but the sun’s surface isn’t hot enough to emit much far UV; that predominantly comes from the ultra hot gas particles above the sun’s visible surface, heated by magnetic effects to millions of degrees. There’s huge variability in those processes, with up to an order of magnitude variation in far UV output, which causes Earths thermosphere to more than double in thickness.
With climate change, the portion of the atmosphere that’s warming only extends to the boundary between the troposphere and stratosphere around 20 km. The gas above that point, in the stratosphere, is actually getting slightly cooler due to greenhouse gasses insulating the stratosphere from the Earth’s warmth (if you put on a blanket, you feel warmer, but you look cooler on a thermal camera). The lower 20 km of the atmosphere is expanding slightly, but a 1 to 2 temperature increase is less than a 1% increase in gas volume, so we are talking about a change on the order of 0.1 km, and this is negated by the shrinking of the next 30 km of stratosphere, such that the net effect of climate change is to shrink the thickness of the atmosphere up through the stratosphere. More importantly, these are very tiny effects from the perspective of the thermosphere, which can double in thickness in response to changes in solar weather, a dynamic range that’s at least 5000x more significant.
Last edited:
Upvote
95
(95
/
0)
theradicalmoderate
Ars Tribunus Militum
Aren't all engines, other than cold-gas thrusters, plasma engines?
It's OK to say "Hall-Effect thrusters". Nobody but us geeks here...
Upvote
22
(22
/
0)