• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Data center in Space

“Launching the idea of data centers in space​

Critical technical aspects of such operations need to be resolved, however, particularly damage to the orbiting data centers from high levels of radiation and extreme temperatures, and the danger of them being hit by space junk.

Another question is how malfunctioning or damaged gear would get fixed in an economical way.

Phillip Metzger, a professor in the University of Florida physics department and a former NASA scientist, reasoned in a recent online post that orbiting data center maintenance could be managed in ways such as using robots and small modular parts that could easily be replaced.

"A lot of the skepticism of data centers in space probably comes from failing to price in the effects of exponential expansion," Metzger said in a recent post on X, formerly Twitter.

"If AI doesn't grow exponentially, then I don't think it will make sense very soon to put them in space; but I think it will grow exponentially."“

Source: https://phys.org/news/2026-02-idea-centers-space.html
 
We have thousands of satellites and they all use batteries and solar, come on they could figure out a way.

No EPA, DEP Army corp, No water, not dealing with power companies, cities, counties, state BS, More sunlight, more power from solar outside the atmosphere, all the cooling you need, perfect vacuum, no land needed.
Elon just merged XAi and spaceX together. He will figure it out.

https://www.spacex.com/updates#xai-joins-spacex
 
Last edited:
sram said:
https://www.eweek.com/news/spacex-a..._86b8chtp8&utm_term=daily-tech-insider-active

SpaceX buys xAI
Click to expand...

"Musk Predicts SpaceX Will Launch More AI Compute Per Year Than the Cumulative Total on Earth (substack.com)34​

Posted by msmash on Thursday February 05, 2026 @03:00PM from the pushing-the-limits dept.
Elon Musk told podcast host Dwarkesh Patel and Stripe co-founder John Collison that space will become the most economically compelling location for AI data centers in less than 36 months, a prediction rooted not in some exotic technical breakthrough but in the basic math of electricity supply: chip output is growing exponentially, and electrical output outside China is essentially flat.

Solar panels in orbit generate roughly five times the power they do on the ground because there is no day-night cycle, no cloud cover, no atmospheric loss, and no atmosphere-related energy reduction. The system economics are even more favorable because space-based operations eliminate the need for batteries entirely, making the effective cost roughly 10 times cheaper than terrestrial solar, Musk said. The terrestrial bottleneck is already real.

Musk said powering 330,000 Nvidia GB300 chips -- once you account for networking hardware, storage, peak cooling on the hottest day of the year, and reserve margin for generator servicing -- requires roughly a gigawatt at the generation level. Gas turbines are sold out through 2030, and the limiting factor is the casting of turbine vanes and blades, a process handled by just three companies worldwide.

Five years from now, Musk predicted, SpaceX will launch and operate more AI compute annually than the cumulative total on Earth, expecting at least a few hundred gigawatts per year in space. Patel estimated that 100 gigawatts alone would require on the order of 10,000 Starship launches per year, a figure Musk affirmed. SpaceX is gearing up for 10,000 launches a year, Musk said, and possibly 20,000 to 30,000."
 
mullet said:
We have thousands of satellites and they all use batteries and solar, come on they could figure out a way.

No EPA, DEP Army corp, No water, not dealing with power companies, cities, counties, state BS, More sunlight, more power from solar outside the atmosphere, all the cooling you need, perfect vacuum, no land needed.
Elon just merged XAi and spaceX together. He will figure it out.
Click to expand...

Vacuum makes is very complicated to remove heat. The big reddish brown "solar panels" on the ISS are actually radiators, the ISS is a pretty low power structure. Data centres need to remove massive amounts of waste heat and space isn't so much cold as it is empty, and there's a reason your fancy travel mug is vacuum insulated.
 
toast0 said:
There's no way it's cost effective. Remote hands is gonna cost a fortune.

Seriously though, power is hard, cooling is hard, networking is hard, maintenance is impossible, and the thing will burn up in a few years. More bitflips up there, too. At least you can get a couple bucks from liquidating 3-5 year old servers on earth.

I guess data destruction is easier though.
Click to expand...
I can think of several IT dudes in their 30's , no wife, no kids, no gf's, no hope, that would go up there to work!
 
Space is cold as eff...working in SI&T at Space Systems Loral (which had Skynet in case you were wondering), the most dangerous test we did was thermal vacuum testing because if we cool too quickly it cracks the bird. And we have had satellites that break in space when the thermal pads fail and it gets too cold, whole panel cracks and the controls end up failing to keep pose because the dynamics completely changed. Cooling with black body radiation isn't an issue. This is well established physics.

There are computers running in space now that are radiation hardened and I think that is easily the biggest risk here for the compute. If the machines are just running inference, no big deal if a bit flip happens there. I assume they can set them up for quick reboots...the flight control computers should still be RAD certified and fully redundant, but that is a separate problem.

Radiation formulas for those that want to get into the physics...

http://www.hyperphysics.phy-astr.gsu.edu/hbase/thermo/stefan2.html#c1

Calculator...
http://www.hyperphysics.phy-astr.gsu.edu/hbase/thermo/stefan.html#c3
 
Last edited:
1.1.2.3.5... said:
Space is cold as eff...working in SI&T at Space Systems Loral (which had Skynet in case you were wondering), the most dangerous test we did was thermal vacuum testing because if we cool too quickly it cracks the bird. And we have had satellites that break in space when the thermal pads fail and it gets too cold, whole panel cracks and the controls end up failing to keep pose because the dynamics completely changed. Cooling with black body radiation isn't an issue. This is well established physics.

There are computers running in space now that are radiation hardened and I think that is easily the biggest risk here for the compute. If the machines are just running inference, no big deal if a bit flip happens there. I assume they can set them up for quick reboots...the flight control computers should still be RAD certified and fully redundant, but that is a separate problem
Click to expand...

So, at 2.7k the 5MW data centres proposed right now would need about 12000 square meters of radiating surface, not a massive deal 3 acres is nothing in space. More importantly it weighs nothing, hell radiators aren't even that heavy to launch.

My thought runs to where these things could orbit, I would think they would have to be in a low orbit similar to StarLink satellites. I wonder how much manoeuvring you would have to do to avoid collisions in such a crowded area. Or could they orbit further out? I know Slop generation isn't as sensitive to latency as remote access or gaming or whatever. We're having to deorbit shit in that area all the time because they run out of propellant for evasive manoeuvres to avoid collisions. You obviously have more information on this than I do, do we even know where these things would be orbiting?

There's a lot of logistics involved, I suspect VAST amounts of tax-payer money and private equity would be needed.

Nuclear on earth is still a better option, we just need to update our risk mathematics and stop trying to regulate nuclear out of existence. We still calculate radiation exposure using a method that has been proven to be completely wrong.
 
Twisted Kidney said:
So, at 2.7k the 5MW data centres proposed right now would need about 12000 square meters of radiating surface, not a massive deal 3 acres is nothing in space. More importantly it weighs nothing, hell radiators aren't even that heavy to launch.

My thought runs to where these things could orbit, I would think they would have to be in a low orbit similar to StarLink satellites. I wonder how much manoeuvring you would have to do to avoid collisions in such a crowded area. Or could they orbit further out? I know Slop generation isn't as sensitive to latency as remote access or gaming or whatever. We're having to deorbit shit in that area all the time because they run out of propellant for evasive manoeuvres to avoid collisions. You obviously have more information on this than I do, do we even know where these things would be orbiting?

There's a lot of logistics involved, I suspect VAST amounts of tax-payer money and private equity would be needed.

Nuclear on earth is still a better option, we just need to update our risk mathematics and stop trying to regulate nuclear out of existence. We still calculate radiation exposure using a method that has been proven to be completely wrong.
Click to expand...
I’m with you on nuclear being the better option, no argument there. Musk tends to chase difficult, unconventional ideas, and this might tie into his broader Mars ambitions. Hard to say what motivates billionaires at that level.

On the orbit question, you’re right that it gets complicated. They could go to geostationary orbit, but the latency is rough. GEO is about 22,500 miles up, and for something like a sat‑phone call the signal goes up, down to a ground station, across the network, then back up and down again. That’s four legs. Round it to 25,000 miles per leg and you’re looking at roughly 100,000 miles of travel. With light speed around 186,000 miles per second (call it 200,000 for easy math), you end up near half a second of latency. We can get more precise if anyone wants to, but the point stands: GEO is slow.

LEO gives you more flexibility because it isn’t a single altitude, it’s a whole band. My background was mostly in integration and test, so I never worked the orbital‑mechanics side directly, but I know ground‑station geometry and coverage play a big role in choosing an orbit. And yeah, LEO is crowded, but there are still workable shells if you’re willing to burn propellant for avoidance.

Big picture, we need more cheap power regardless of whether space‑based data centers ever happen. Solid‑state batteries are coming, EV adoption will accelerate, and the grid will feel that pressure. Nuclear is the obvious answer, but neither political party seems willing to touch it, even though it’s one of the few issues that could get broad bipartisan support. I wish I understood that reluctance better.
 
They already know that for it to be feasible they have to get the launch costs down to $200 per kilogram. it's $1000 or more currently.

At some point it likely will be feasible, but probably a decade off. If they design it such that robots can replace parts, I bet they can make it work. SpaceX has some pretty brilliant people working there, see Falcon9. (but don't look at Starship...:)
 
It's getting to the point where Elon will have to shave his head and wear a silver shiny suit.
 
erek said:

“China joins race to develop space-based data centers with 5-year plan​


U.S. companies also plan to build solar-powered data centers in space as the market becomes crowded on Earth.”

https://www.space.com/space-explora...lop-space-based-data-centers-with-5-year-plan
Click to expand...
Welp, perfect timing, I was just about to comment.

The only way I see this taking off is if China decides to look into it. The US will spare no expense to stay ahead of China in the space race.
 
Last edited:
  • Like
Reactions: erek
like this
Twisted Kidney said:
My thought runs to where these things could orbit, I would think they would have to be in a low orbit similar to StarLink satellites. I wonder how much manoeuvring you would have to do to avoid collisions in such a crowded area. Or could they orbit further out? I know Slop generation isn't as sensitive to latency as remote access or gaming or whatever. We're having to deorbit shit in that area all the time because they run out of propellant for evasive manoeuvres to avoid collisions

There's a lot of logistics involved, I suspect VAST amounts of tax-payer money and private equity would be needed.

Nuclear on earth is still a better option, we just need to update our risk mathematics and stop trying to regulate nuclear out of existence. We still calculate radiation exposure using a method that has been proven to be completely wrong.
Click to expand...
This is my concern as well, LEO is getting increasingly congested and we plan to launch more and more satellites especially for military use (see the PWSA). It isn't just us either, China has been flooding space recently with their own satellites and Iran has got stuff going on as well.

I wonder how close we are for the Kessler Syndrome to become a real concern. This is a scenario in which our orbital space is so congested that a single event will cause a domino effect, creating a cloud of debris that expands and basically destroys everything in orbit. It would take decades to clean up before orbital space would be usable again. Think we still got debris back from when China was testing their ASAT missiles in 2007. Tiny pieces of debris are still capable of ripping through anything they come in contact with due to the speeds of everything up there.

This is also why cybersecurity is so important. Using physical weapons against satellites will be too risky due to the chain reactions it can cause; much better to disable them through cyber means. Also gives you plausible deniability. A lot of the older stuff just relies on security through obscurity. Imagine the damage a terrorist could cause by hacking a satellite and causing a collision.

I'm wary about going the nuclear on earth route; honestly before I was a huge supported of nuclear, but as I've learned more about cybersecurity and seeing how behind the US is in it, I'd rather not. I'd rather screw up our orbits than screw up our planet if something goes wrong.

Really interesting stuff
 
Last edited:
  • Like
Reactions: erek
like this

“Voyager Technologies CEO says space data center cooling problem still needs to be solved​


KEY POINTS
  • Voyager Technologies CEO Dylan Taylor said data centers in space will become a reality, but cooling issues still need to be resolved.
  • Tesla CEO Elon Musk highlighted space-based data centers as a major reason for merging SpaceX and xAI.
  • SpaceX’s potential IPO this year and President Donald Trump’s focus on defense spending have lifted interest in the space sector.”
https://www.cnbc.com/2026/02/06/voyager-technologies-cooling-space-data-centers.html
 
“The high water demand of data centers causes significant problems. Large volumes of often potable, drinking-quality water are used for cooling, putting pressure on local water supplies and competing with households and agriculture. One way around this is to do what China has done — place data centers in a location where they're literally surrounded by non-potable water — the ocean. There have been several experimental underwater projects — like Microsoft's Project Natick, which concluded in 2020 — but currently, the only operational, commercial underwater data center in the world is in Hainan Province in China.

Over the past few months, this data center has begun operating as a large-scale artificial-intelligence computing facility, like DC's "AI Alley". Sealed inside steel capsules placed on the seabed near Lingshui, rows of servers now perform the same kinds of tasks as those in conventional data centers — processing data, running cloud services, and training AI systems — but are cooled naturally by surrounding seawater rather than by energy-intensive air-conditioning. This recent shift toward high-density AI computing marks the most advanced stage of a project that aims to combine computing power with lower energy use and a smaller land footprint.

The facility itself is not new. The first underwater modules were deployed in 2022 and entered commercial operation in 2023. The additional modules and upgraded hardware enabled it to take on far more demanding AI-focused tasks, transforming an already operational facility into a full-scale computing hub. China has another underwater data center – still under construction — near Shanghai.”

Read More: https://www.slashgear.com/2094111/china-underwater-data-center-ecosystem-study/
 
“Terrestrial data centers are so 2025. We're taking our large-scale compute infrastructure into orbit, baby! Or at least, that's what Big Tech is yelling from the rooftops at the moment. It's quite a bonkers idea that's hoovering up money and mindspace, so let's unpack what it's all about – and whether it's even grounded in reality.


Let's start with the basics. You might already know that a data center is essentially a large warehouse filled with thousands of servers that run 24/7.


AI companies like Anthropic, OpenAI, and Google use data centers in two main ways:

  • Training AI models – This is incredibly compute-intensive. Training a model like the ones powering OpenAI's ChatGPT or Anthropic's Claude required running calculations across thousands of specialized chips (GPUs) simultaneously for weeks or months.
  • Running AI services – When you converse with those models' chatbots, your messages go to a data center where servers process it and send back the model's response. Multiply that by millions of users having conversations simultaneously, and you need enormous computing power ready on-demand.
AI companies need data centers because they provide the coordinated power of thousands of machines working in tandem on these functions, plus the infrastructure to keep them running reliably around the clock.

To that end, these facilities are always online with ultra-fast internet connections, and they have vast cooling systems to keep those servers running at peak performance levels. All this requires a lot of power, which puts a strain on the grid and squeezes local resources.

So what's this noise about data centers in space? The idea's been bandied about for a while now as a vastly better alternative that can harness infinitely abundant solar energy and radiative cooling hundreds of miles above the ground in low Earth orbit.”

https://newatlas.com/ai-humanoids/space-based-data-centers-ai-explained/
 
erek said:

”Why the economics of orbital AI are so brutal​


In a sense, this whole thing was inevitable. Elon Musk and his coterie have been talking about AI in space for years — mainly in the context of Iain Banks’ science-fiction series about a far-future universe where sentient spaceships roam and control the galaxy.”

https://techcrunch.com/2026/02/11/why-the-economics-of-orbital-ai-are-so-brutal/
Click to expand...
He has also been talking about Mars for years...and let us not fo get Boring.
 
littlegreenmenfromhome said:
He has also been talking about Mars for years...and let us not fo get Boring.
Click to expand...
No kidding. Let's leave aside whatever your personal feelings on Musk are, everyone should be able to admit that it is an objective fact that just because he says something, doesn't mean he's going to make it happen. Manned mission (or anything) to Mars, Boring Company, Hyperloop, none of these have delivered on their promises or are anywhere near delivering. The Hyperloop is straight up abandoned near as I know. Boring Company I think still exists but he found out the only way he can make tunnels cheaper is by making them smaller, shorter, less complex, you know all the normal shit that raises costs. Space X doesn't even have the tech to send a basic payload to Mars, much less having actually done so.

Just because Musk talks about something and even puts (investor) money behind it doesn't mean it'll happen.
 

View: https://www.youtube.com/watch?v=BYXbuik3dgA

In this interview from a couple weeks ago he talks about data centers in space a bunch.

To summarize it:

Building power plants on earth is the bottleneck for AI datacenters. And it is so slow on earth that he thinks it will be faster to build them in space.

He thinks it will only take 5 years before they start deploying them. IMO an extremely overoptimistic and probably impossible timeline which is typical with Musk.
 
Sycraft said:
, everyone should be able to admit that it is an objective fact that just because he says something, doesn't mean he's going to make it happen. Manned mission (or anything) to Mars, Boring Company, Hyperloop, none of these have delivered on their promises or are anywhere near delivering. The Hyperloop is straight up abandoned near as I know. Boring Company I think still exists but he found out the only way he can make tunnels cheaper is by making them smaller, shorter, less complex, you know all the normal shit that raises costs. Space X doesn't even have the tech to send a basic payload to Mars, much less having actually done so.

Just because Musk talks about something and even puts (investor) money behind it doesn't mean it'll happen.
Click to expand...
Oh no doubt, at one point someone referred to him as "the real life Tony Stark" and I think he literally took that to heart. Not saying he hasn't done anything good, he has, but just because you have something that works well, doesn't mean everything you come up with will also. His Boring Company and Hyperloop are nothing more than schemes he's used to sell more cars, he's even admitted his hair brained scheme for the Hyperloop was to derail (pun intended) California's High Speed Rail project, because he didn't want people taking the train anywhere, instead wants them in his cars, and while it probably would not have quite the effect it has had in Europe or Japan, he tried to stop it any way he could, little did he know there were plenty of others who were quite a bit more successful at stopping it (even though it's still under construction), and it was an idea he stole from some idea from around 100 years ago. This is why he so quickly gave "his idea" to the public to do with it what they want, which is absolutely perfect for him, because not only is he no longer responsible for a money pit tech that won't work, but if by some reason it does work... he gets to claim it was his idea.

Ditto with the Boring company, to come from the dumb idea that more lanes of on the freeway really doesn't do much of anything to alleviate traffic, he has an idea to essentially "build more lanes" but do so in the most expensive way possible, underground! Of course this will be Tesla only tunnels, and I wouldn't be surprised if he got this idea from the Simpsons (Stonecutters episode). Now his plans are to stop making Teslas at California plants, and instead make humanoid robots... yeah... not sure if that's a jab directly at California (which he has done plenty), but I wouldn't be surprised if the Boring Company slowly evaporates from his portfolio as well.

p.s. I stopped liking the man after that Hyperloop nonsense, he can support whatever political party he wants.
 

“Could AI Data Centers Be Moved to Outer Space?​

Massive data centers for generative AI are bad for the Earth. How about launching them into orbit?

DATA CENTERS ARE being built at a frantic pace all over the world, driven by the AI boom. These facilities consume staggering amounts of electricity. By 2028, AI servers alone may use as much energy as 22 percent of US households. Of course that demand will raise energy prices for everyone, and we’ll need more power plants, which means more global warming.“

https://www.wired.com/story/could-we-put-ai-data-centers-in-space/
 
erek said:

“Could AI Data Centers Be Moved to Outer Space?​

Massive data centers for generative AI are bad for the Earth. How about launching them into orbit?

DATA CENTERS ARE being built at a frantic pace all over the world, driven by the AI boom. These facilities consume staggering amounts of electricity. By 2028, AI servers alone may use as much energy as 22 percent of US households. Of course that demand will raise energy prices for everyone, and we’ll need more power plants, which means more global warming.“

https://www.wired.com/story/could-we-put-ai-data-centers-in-space/
Click to expand...
Meta here in Denmark are using the same power as 30% of our population already in their datacenter FYI.
 
erek said:
These facilities consume staggering amounts of electricity. By 2028, AI servers alone may use as much energy as 22 percent of US households.
Click to expand...
That line right there is one of the problems: How do you generate that much energy in space and if you have something that does a good job of it, why couldn't the same technology simply be used on Earth for less money? I mean maybe there's something I'm missing, but any technology I can think of either won't work in space or will work just as well on Earth. Like any fuel-based tech is out. Sending fuel to space is stupidly fucking expensive, it is one of the big issues with any kind of return mission to places like Mars. Solar works in space... but of course solar works great on the Earth too. Now if you go up to geostationary orbit you can be in a position where solar is better, since you can be in the sun 24/7. However, I can't imagine they are interested in that, latency would be killer. I'm assuming LEO is the target... In which case you will be going in to darkness regularly and so need batteries in addition to the panels, just like you do on Earth.

I'm just not aware of anything that is good at generating power in low Earth orbit that is not also good at generating power on the surface of the Earth, where you don't have to deal with launch costs and you can repair it. The problem with solar for datacenters isn't that it won't work, it is that techbros don't want to pay for it. It's cheaper to just use the grid and make everyone pay more, or to use on site fuel generators. There's nothing stopping them from building big solar and battery arrays, other than them not wanting to.
 
Sycraft said:
I'm just not aware of anything that is good at generating power in low Earth orbit that is not also good at generating power on the surface of the Earth
Click to expand...

Solar is space is better than on Earth - sunlight 24/7 and you don't have Earth's atmosphere scattering up to 50% of the sunlight before it hits your panels.
 
kram182 said:
Solar is space is better than on Earth - sunlight 24/7 and you don't have Earth's atmosphere scattering up to 50% of the sunlight before it hits your panels.
Click to expand...
It's not 24/7 if you are in Low Earth Orbit, which is where I would assume these things are going to go. There you are moving around the Earth quite quickly, you go from Sun to shadow all the time, hence the need for batteries. As for atmospheric scattering, I'd be interested to see a chart of how it effects output to see what the difference is.

That said, there's still the small issue of the $1000 per kilogram (or more) of space launch. So assuming really good 30% efficient panels and that solar output number I found for out-atmosphere is correct, you are looking at a peak power output of about 400W/m^2. Assuming these are on the lighter side, that's 10kg so $10,000 to launch 400 watts...

...That uh... buys a lot more panels and batteries here on Earth.
 
Sycraft said:
It's not 24/7 if you are in Low Earth Orbit
Click to expand...

Yes it is - not 100% exactly but much closer to it than you're thinking. 'Sun-Synchronous Orbits' is the specific name for it (staying in sunlight as much as possible while in LEO)

Sycraft said:
As for atmospheric scattering, I'd be interested to see a chart of how it effects output to see what the difference is.
Click to expand...

Without even getting into the exact physics of it (you can look it up) - atmosphere is a soup. It's easier to get anything (even light rays/photons) on a direct line of sight shot through 'nothing' than 'a soup' - especially when you then add the water/moisture/soup particles refract light etc.

Sycraft said:
That said, there's still the small issue of the $1000 per kilogram (or more) of space launch. So assuming really good 30% efficient panels and that solar output number I found for out-atmosphere is correct, you are looking at a peak power output of about 400W/m^2. Assuming these are on the lighter side, that's 10kg so $10,000 to launch 400 watts...

...That uh... buys a lot more panels and batteries here on Earth.
Click to expand...

As with anything the way to get better at something/get costs down is to keep at it/doing it.
 
kram182 said:
Without even getting into the exact physics of it (you can look it up) - atmosphere is a soup. It's easier to get anything (even light rays/photons) on a direct line of sight shot through 'nothing' than 'a soup' - especially when you then add the water/moisture/soup particles refract light etc.
Click to expand...
I don't know the physics of it, but looking up numbers what I found that solar irradiance in space was about 1.36kW/m^2, as opposed to about 1kW on the surface. So about a third more.

kram182 said:
As with anything the way to get better at something/get costs down is to keep at it/doing it.
Click to expand...
Have they made new engines that use less fuel? That's the alpha and omega of space launch cost. The only way you are making significant improvements to cost is to either design an engine that delivers significantly more thrust with a given amount of fuel usage, or switching to a new fuel.

The idea that just "doing more of it" will bring down the cost much is false. In manufacturing yes, that applies (to a point) since there's a heavy fixed cost in setting up a factory/production line and much less cost in operating it once going so the more you produce the more units to spread that cost over. With space launch, not so much. The big cost is fuel and the amount of fuel you need is dictated by the rocket equation. The only ways to bring it down are to carry less weight, get more efficient engines, or change to a different kind of fuel.
 
Sycraft said:
I don't know the physics of it, but looking up numbers what I found that solar irradiance in space was about 1.36kW/m^2, as opposed to about 1kW on the surface. So about a third more.
Click to expand...

It's the same reason we have our best telescopes in space - to avoid the light rays scattering away/being obscured/etc- this same stuff applies to solar panels - doesn't matter if trying to absorb the light for sight or power.

Sycraft said:
Have they made new engines that use less fuel? That's the alpha and omega of space launch cost. The only way you are making significant improvements to cost is to either design an engine that delivers significantly more thrust with a given amount of fuel usage, or switching to a new fuel.

The idea that just "doing more of it" will bring down the cost much is false. In manufacturing yes, that applies (to a point) since there's a heavy fixed cost in setting up a factory/production line and much less cost in operating it once going so the more you produce the more units to spread that cost over. With space launch, not so much. The big cost is fuel and the amount of fuel you need is dictated by the rocket equation. The only ways to bring it down are to carry less weight, get more efficient engines, or change to a different kind of fuel.
Click to expand...

You wouldn't be seeing the push for it you are if it weren't true - we've been developing methane/reusable rockets specifically for stuff like this/increase or decrease of cost/efficiency/etc. Again, the only way to get better/if at first you don't succeed.....

Edit: Also, solar panels suffer more risks on Earth than space (I didn't say 'none' in space) just due to more 'damaging variables' regardless of how difficult space is for us humans/repair/etc
 
Last edited:
kram182 said:
Again, the only way to get better/if at first you don't succeed.....
Click to expand...

If the projected costs are wildly beyond the projected value, refinement doesn't help. Making more pennies never got cheaper enough to make up for the cost of the raw materials. Same deal with rocket fuel --- some fuels may be relatively less expensive, but not like 10% of current cost.

Fabbing in space might happen eventually for the couple of things where the complex setup adds enough value to justify it. Datacenters of size in space just doesn't make sense. There's plenty of room on the ground, and it's nice to be able to do maintenance, whether that's replacing flakey cables or replacing racks with better hardware. Could starlink include some SSDs and run a CDN in space? Yeah, that might work, probably not terribly helpful for users that are one bounce to the teleport, but could be handy for users where traffic goes through multiple satellites before getting back to the ground.
 
toast0 said:
If the projected costs are wildly beyond the projected value, refinement doesn't help.
Click to expand...

This is again is falsely assuming costs/technology/capabilities are fixed and stagnate over time

History shows the trend is cost down technology/capability is up over time, regardless of field/sector
 
Sycraft said:
I don't know the physics of it, but looking up numbers what I found that solar irradiance in space was about 1.36kW/m^2, as opposed to about 1kW on the surface. So about a third more.
Click to expand...
I do know the physics of it, and it's a lot more complicated than simply looking at raw power numbers since solar panels don't use each wavelength equally as far as power output, i.e. a blue photon (which is more likely to be scattered by our atmosphere) will knock an electron loose in a panel but the power output of that electron is strictly because it got knocked loose, any further energy the blue photon added gets wasted as heat, so a red photon which is less likely to be blocked will still make the same amount of power as a blue (in fact more, because that wasted heat lowers the efficiency). Not to mention most of the sunlight we receive is infrared, and our atmosphere does block some chunks of it, and those longer infrared bands really isn't particularly useful for solar power, and to a lesser extent UV radiation as well. While yes, solar panels do make more light in space on a per meter² per second basis, the fact that they can be positioned to be in the sunlight for far longer than they can on the ground is a more beneficial aspect of them being in space.

And yes, there are different formulae for making solar panels, of which I'll admit I'm not up to date on the latest state of the art, but a quick google shows this for silicon based ones, and you can see in those higher energy colors (the ones that our atmosphere largely blocks) silicon has a lot more wasted energy from them, even though they are represented as stronger in a W/m² value.

Si-Absorption-e1581357321237-1024x653.png
 
You must log in or register to reply here.
Back
Top