Starlink quietly lost over $250M in burned satellites this summer

[modeless]

Lost 200 but launched 345, almost all of which are the latest upgraded version. The satellites are designed to last 5 years (the lifetime is limited by propellant because of atmospheric drag at their low altitude) and the first batches were launched 4 years ago now. As the constellation matures they will be de-orbiting satellites about as fast as they launch them.

Seems to me like the long term viability of Starlink (if it's not already profitable) will be assured if Starship meets its goals. They'll be able to launch way more satellites at once, for less per launch, and with faster turnaround.

[usrusr]

Launch way more, and I'd imagine the sweet spot for the amount of fuel each satellite gets to shift up considerably: with a launch the size of starship they can saturate the orbits reasonably reachable from that launch for decades (spares remain in higher orbits with less drag I think) and giving one satellite more fuel is cheaper than deploying two satellites with less fuel each. And achieve far more than twice the runtime, because the fuel complement in today's starlinklings is certainly quite far from 100% mass fraction. The current compromise is far quick global reach running on F9, with a generous dose of iteration.

[raesene9]

What's not clear from the article is whether the ones lost were old ones or newer ones.

If they're losing ones that were close to their max. life anyway that's possibly no great loss, however if they were newer satellites on top of expected attrition that's more of a concern.

[threatripper]

How many satellites can be launched at once? I assume that the number of satellites per orbit will be limited and changing the orbit would take extra fuel that either the launch vehicle or the satellite has to provide.

[mlindner]

It's unfortunate that this misinformation is being upvoted on this site. The headline is completely incorrect, as is the statistics they're using to base the article on.

https://twitter.com/Marco_Langbroek/status/17055628292254106...

https://planet4589.org/space/con/star/stats.html

There has only ever been 350 starlink satellites that have deorbited, and only 8 within the last two months.

[justapassenger]

> Seems to me like the long term viability of Starlink (if it's not already profitable) will be assured if Starship meets its goals. They'll be able to launch way more satellites at once, for less per launch, and with faster turnaround.

This is just wishful thinking without any data backing it up.

[npunt]

Lost $250M using the advertised launch price of $67M for Falcon 9, which of course is way higher than what it actually costs SpaceX.

[pbmonster]

...and completely ignoring cost of lost hardware. Aren't those satellites several $100k a piece?

[Rebelgecko]

There's reason to be skeptical of that. I think their most recent profitable quarter (maybe their only profitable quarter?) was with a profit margin of ~3%.

Granted maybe there's some marginal benefits, eg launching n+1 rockets is cheaper per rocket than launching n rockets

[fooker]

The fuel costs about half a million.

Beyond that it's mostly people's times and materials they have stockpiled for economies of scale.

I would guess the lost satellites cost a bit more to manufacture that to launch in a reused Falcon 9.

[m_mueller]

does it? why can they then sell it for not much more and run a profit?

[codegladiator]

So we have reached that stage of Universal Paperclips

[jefozabuss]

Be careful googling this if you have a weak resistance to incremental games like I do as some hours will fly by quicky...

[helsinkiandrew]

That's about a 4.4% loss (200 of 4500 satelites which are expected to have a life of 5 years), and the internal cost to Starlink of a new launch is probably far less than the $67 million 'retail' cost of a SpaceX launch.

[vardump]

I think they're prepared to lose satellites, that's the nature of large swarms.

Some losses might be intentional for safety after damage or low propellant. They really can't afford to put any debris on their orbits.

[kvdveer]

IIRC debris is not a major concern in the orbits Starlink is using. There is a non-trivial amount of atmospheric drag. This will slow down anything without active propulsion, causing them to drop from orbit, and eventually burn up.

[eunos]

Speaking of Starlink, what piqued my interest is that Huawei Mate 60 incorporated satellite call module without being bulky. It is not yet able to do data transfer but in case they can, isnt that practically a portable starlink equivalent?

[chrismorgan]

Starlink is low earth orbit, around 550km up, adding best-case latency of under 4ms (550km*2/c ≈ 3.67ms). That allows an excellent service.

Satellite phones (and I’ve confirmed this applies to the Huawei Mate 60) connect with satellites in geostationary orbits, meaning an altitude of about 35,786km, adding latency of around 240ms (35786km*2/c ≈ 238.74ms). This can be acceptable for some purposes, but harms things far more than you might imagine. (Source: personal experience with the Australian NBN SkyMesh or whatever it was called before NBN took it over, only once, about five years back. Look, a lot of the internet is located in the USA, and many things perform vastly worse from Australia than from the USA, e.g. several second page loads instead of under one second, and this satellite connection was basically that but even for stuff in Australia, and several times as bad for stuff in America.)

[cf141q5325]

Thanks for mentioning that. Its the pro version and it apparently entails voice and text. Which makes it sound more like satellite phones then internet over satellite. Still, quite the interesting feature.

[inemesitaffia]

More like Iridium

[grouseway]

Is this because they are making a trade-off between surving some amount of solar flare activity vs increasing satellite weight by hardening them to radation?

I imagine it's preventable since the ISS and other large orbiters stay functional forever.

[bagels]

Is this intentional? Is it just happening for newly launched satellites?

[vasco]

If you read the article you can answer both of your questions. The Sun has no intentions, and solar flares don't choose cohorts of satellites.

[Ekaros]

Going forward what is the expected yearly attrition rate? So how much have they spend a year to keep replacing the natural losses of their satellites?

[mlindner]

It's unfortunate that this misinformation is being upvoted on this site. The headline is completely incorrect, as is the statistics they're using to base the article on.

https://twitter.com/Marco_Langbroek/status/17055628292254106...

https://planet4589.org/space/con/star/stats.html

There has only ever been 350 starlink satellites that have deorbited, and only 8 within the last two months.