Are we ready to see satellites built on an assembly line approach? | Hacker News Reader

Are we ready to see satellites built on an assembly line approach?

blog.satsearch.co

45 pointscosmosguru44 comments

With the number of satellites slated to go up, there will be enough shade to cool the earth and save us from climate change.

I know you're not being serious, but I like estimating stuff like this.

The biggest solar panel array in space right now, AFAIK, is the ISS at 2500 m^2.

There are currently ~6000 satellites in (any) orbit. Starlink has 12000 more planned. Let's go nuts and say there's gonna be 50000 total satellites.

Let's say ALL of them are as big as the ISS's, ridiculous as that is. And let's say they cast a 1:1 shadow, even though that's also patently ridiculous. I'm just making all of these number as big as possible so nobody can possibly say "no, it's bigger".

2500 m^2 * 50000 = 125 square kilometers.

So with many times the number of satellites planned, with each one being anywhere between 50 and 100 times as big as it really is, with each one casting a shadow as big as it is... we shade - out of the entire planet - an area about 1.5 the size of Manhattan.

You don't need to cover the whole planet, just the parts under the sun. I'm not an orbital machinist, but maybe we could park just enough blockers in an orbit that's always between the sun and Earth to halt the warming. How much of the sun would need to be blocked to bend the curve down? Everything necessary on the ground to stop putting CO2 in the air is already underway. All we need is more time.

This is something that's been considered, particularly for doing things like terraforming Venus. There are these areas called Lagrange points[1] that provide a location where one can park a satellite where it'll remain stationary relative to the two-body system of interest. The point directly between the sun and earth is the L1 point and allows you to place a sunshade such that it'll always be directly between the sun and earth. It should be noted, however, that these points are generally not stable and so require some degree of station-keeping to remain viable long term.

[1] https://en.wikipedia.org/wiki/Lagrange_point

A solar shade would also be a solar sail -- it could maneuver by changing its shape to catch the solar wind differently. It would still need to station-keep (perhaps even more because of fluctuations in solar wind), but it wouldn't need to use propellant.

Can you tack in space?

There has been some research on this: https://en.wikipedia.org/wiki/Space_sunshade

It's interesting how cheap it can be done ($20B-$200B), which may be $1T in real life, but that's still cheap compared to the programs the governments are running.

The moon has the perfect size already. We just need to nudge it around a little bit to provide some local shade. Solar eclipse on demand, how hard can it be?

You know, you might be on to something. Set up a mine to blast rock out from the moon to create a ring at a higher orbit to block the light. The moon will eventually scoop it all back up as it continues its trip away from the planet.

With 1,300 Starlink satellites in orbit, and many thousands scheduled to go up, I think we're already seeing the assembly line approach to building satellites.

There are favorites coming up in Europe, China, India and Japan apart from the US.

It almost seems like taking a leaf out of the auto industry. The question is, how many of these can survive eventually!

Very few, capital cost to build and launch is quite significant, cost to operate is significantly less, but still substantial. You need a lot of subscribers to make this worthwhile and this is likely will be winner take most tech market. Left alone global LEO market would probably support 1-2 constellations. Given geopolitical nature it is likely that some constellations will be outlawed in certain areas. SpaceX will likely capture a 75%+ portion of the market everywhere it is not prohibited.

It is feasible to see project Kupier and One Web eventually getting competitive with Starlink 1.0, but they are not launching or innovating fast enough. By the time they are ready to start offering V1 constellation SpaceX will be launching V2 and testing V3. Long term the only real competitor is China because they have a large protected market and capable of subsidizing launch and scaling up incredibly rapidly.

Sorry I can't resist but put my hacker hat on:

How would effective prohibition work? Beams are directed enough to make them hard to find from a frequency perspective. Payments will probably be possible in BitCoins at some point given Teslas move in that direction. Controlling the import of hardware seems like a way to do it, but I bet it would still show up everywhere.

It's easy: Just make it illegal, and put make it incumbent on starlink to enforce it. Violations are met with (eg) sanctions on Tesla and other related parties.

This has played out many, many times in history and it's generally successful.

Yes, it's often not perfect. There are small numbers that slip through the cracks. But as soon as it's any significant number the crack will get closed. c.f. great firewall of china, north korea anything, various European data location requirements, russian internet payments and search, etc etc etc etc.

That's easy. If spacex does not respect the ban, China will fire anti-satellite missiles to the spacex constellation.

That's not easy. There will be tens of thousands of satellites. It would be prohibitively expensive to hit them all with missiles. Not to mention the debris cloud which, while temporary, would cause real problems for everyone in space for a few years.

More feasible attacks would be hacking the satellites or ground systems, RF jamming, or just locating users on the ground by their RF emissions and throwing them in the gulag. Or how about holding the Shanghai Tesla factory hostage? Less fancifully, China can put pressure on SpaceX via the ITU. There are plenty of options for China and there is zero chance SpaceX will try to break Chinese law.

Hitting one satellite is a hell of a warning shot though. Literally no one wants the debris clouds.

Every country, US included, would pressure SpaceX to stop operating in countries that don't want it if it were to result in any debris.

no no

no no no

please. if you want to make Starlink illegal what you do is state Starlink is illegal and maybe make it illegal to pay Starlink. that's a lot cheaper and doesn't piss off the international community.

This URL appears to be specific to 1U through 6U cubesats, which are considerably smaller and less capable than a single starlink satellite. Or Oneweb, the most recent other medium large volume satellite manufacturing project.

Of course neither company is publishing any photos/schematics or details whatsoever about their assembly line, because that's a competitive advantage.

There are multiple pictures of OneWeb´s assembly line in their website, https://onewebsatellites.com/factory/ , including pseudo-blueprints: https://bit.ly/32QSDBS

Still far from what you would expect from a truly automated production line, but it has dedicated stations that allows workers to assemble multiple satellites in parallel.

These smaller satellites can support cameras, low data rate communications like IOT via space, etc.

I guess if you consider space as the 'highway', the size of the satellites can be motorcycles, small cars to big trucks!

I wonder if it will make sense to go back to multi sensor platforms. We could have a fleet of platforms in different altitudes, combining different sensors and comm tech. Might be better than to have a fleet for every specific wavelength (for observation) and radio band.

The problem is how do you do that when 20 companies are flying the same type of satellites and are fighting for customers with almost similar capabilities. If it was a question of science or governmental cooperation, such agreements may be easier.

Anybody know how dead ones are recycled or is it junkyard in space?

"Recycled"? None.

Deorbited: quite a few. Depends on orbital altitude. Low-earth orbit satellites generally deorbit by themselves eventually, but at the upper range of low-earth orbit that can take a few thousand years unless they are intentionally deorbited.

In geosynchronous orbit, nothing is ever deorbited, it's too costly in terms of fuel. They are instead raised in altitude a few hundred klicks to get them out of the way.

There is a junkyard for GEO i.e 360000km orbit. But there are now concepts and companies coming up to solve the junk problems for the 500-2000km orbits - https://www.bbc.com/news/science-environment-56420047

Like any other externality that isn't constrained by regulation, the problem is raging out of control.

https://www.bloomberg.com/opinion/articles/2021-04-17/space-...

random question for anybody that can answer:

for satellites that are scheduled to be "deprecated," can we use a different rocket to gently nudge them out of orbit and have them float off into space? even if you nudge them at something super low velocity like 100MPH into a direction, presumably it would continue floating endlessly at 100MPH until it would hit something?

i'm guessing the building a separate "rocket that nudges dead satellites out of earths orbit" is extremely expensive?

"I'm guessing the building a separate "rocket that nudges dead satellites out of earths orbit" is extremely expensive?"

Yes, at least as expensive as launching the original satellite, if not more (because it has to have the ability to intercept).

Also, just adding a little nudge doesn't really help. It changes the orbit slightly, but it isn't like an additional 44 meters per second (which is what 100 mph is) of velocity is going to make the derelict spacecraft achieve escape velocity. You'll just change the shape of the orbit some. You need to add an additional ~ 3300 m/sec (7400 mph!) in velocity to get to escape velocity.

Decreasing the velocity, such that the derelict eventually de-orbits and burns up in the atmosphere is typically much less expensive. A few hundred m/sec would probably do it, but it depends a lot on what orbit the derelict starts in.

It doesn't work like this. If you nudge something in a circular orbit by 100MPH, in general it moves into a very slightly higher, lower, or more elliptical orbit. For a typical satellite in LEO, the size of the "nudge" you need to get it to "drift away" is comparable to all the rocket fuel that went into getting it there in the first place.

That isn't right, since all you have to do is lower the orbit until it hits the atmosphere, and let that do all the work. A Space Shuttle deorbit burn was ~90 m/s, versus the 7,600 m/s that the ISS moves at. (With even more lost to drag, gravity, etc, getting there.)

The ISS is in a relatively low orbit at ~400km altitude, geostationary satellites are up at 35,000km and have basically zero drag.

If you give a satellite a shove, it will end up on a slightly higher or lower (or eccentric) orbit depending which way you shove. It won’t keep floating off endlessly into space unless you shove it with enough energy to get to escape velocity.

You would deorbit the sat and let it burn up into the atmosphere. In orbit, you can't really "drift off into" space like you're saying here.

I seriously recommend investing in a copy of Kerbal Space Program. There is no better way to gain an intuitive grasp of orbital mechanics.

Are we ready to just see satellites and no stars anymore?

I see a double edged sword here. We already have enough light pollution that stars are hard to see in most cities. That amount of pollution grows yearly as cost per lumen drops. The problem ground based astronomers are having is the pollution these low earth sats are having. This is the down side. What makes me hopeful is how much better orbital optics can be. Couple this with signal processing and scores of telescopes in orbit, and you could create a virtual lens many times larger and clearer than anything planet side.

As the price for getting satellites into orbit drops I imagine astronomy having some pretty major breakthroughs.

That’s at least my hope. Most of what I’ve been watching is low orbit satellites but I don’t know how high an orbit you’d want for the kinds of things I’m imagining.

Something that many in the astronomy community are asking https://www.cnbc.com/2020/04/29/elon-musks-spacex-reducing-s...

The opacity desires of astronomy are different than human eyes. Painting satellites black (ongoing w/ Starlink) is enough to make them dark to the human eye.

I think it is an interesting idea, but I fear that unless we find a way to bring down the broken, disabled, etc. satellites we have, this will lead to accelerating the possibility of a Kessler Syndrome phenomena.

There is a simple and comprehensive solution to this problem which SpaceX has already implemented for Starlink. Simply put all of the satellites under 600 km altitude. Anything at that altitude will fall out of orbit in a few years due to atmospheric drag, unless it continuously boosts itself up. And before you ask: no, collision debris is not ejected into orbits that decay significantly slower than the original orbits, just due to how orbital mechanics work.

>collision debris is not ejected into orbits that decay significantly slower than the original orbits, just due to how orbital mechanics work.

True, but debris can (and has been) thrown into orbits with an apoapsis (high point) much higher than the original collision. Yes, the periapsis is still at the original altitude of the collision but the debris may collide with other satellites or debris creating secondary debris clouds at a much higher altitude than the original collision, and so on.

This point is that the periapsis is still subject to strong atmospheric drag.

And while secondary debris clouds at might higher altitude is theoretically possible, it's vanishingly unlikely due to the vastly lower density.

Reply on news.ycombinator.com