3 ms·

Any idea how they're handling the space data center heat problems? Energy harvesting in space is easy, while getting rid of heat is quite hard.

by zulux 3mo ago

piloto_ciego 3mo ago
Scott Manley had a really good video on it that more or less confirmed my "bar napkin approximations" when people first started talking about this stuff. I think it's this one? https://youtu.be/DCto6UkBJoI?si=opcoaBEqKmw4LCrF https://youtu.be/DCto6UkBJoI?si=opcoaBEqKmw4LCrF Anyway, I think sun synchronous orbit (basically slightly off from polar) and pointing the solar panels toward the sun and the radiator away... It's not as far fetched as people would have you believe.
- JumpCrisscross 3mo ago
  Yup. It’s weirdly closer than I would have thought before sitting down with pen and paper.
- jordanb 3mo ago
  Did we watch the same video? My takeaway from Manley is that it is something that can pencil out as a physics problem with a very carefully selected orbit But as an engineering problem it's quite another thing. For example he talks about how far you have to pump the ammonia, and then you either need a very large amount of ammonia, or you have to pump a smaller amount of ammonia very fast through kilometers of plumbing. Of course there are highly theoretical ideas like making an atmosphere of coolant around the satellite or something but he left that stuff on the sci fi category.
- gamblor956 3mo ago
  That's like saying that controlled fusion is a napkin problem. The issue has always been the engineering. Programmers tend to be extremely bad at estimating he complexity of real world problems.
maxboone 3mo ago
https://www.nextbigfuture.com/2026/02/spacex-has-the-best-radiative-cooling-solution-and-it-is-working.html https://www.nextbigfuture.com/2026/02/spacex-has-the-best-ra...
- zulux 3mo ago
  Fair point, and thanks. Certainly true that Starlink satellites are already successfully dealing with the problem and use a lot of energy. I think the trouble with my thinking was that I was locked into the Shuttle/STS numbers of waste heat transfer.

Any idea how they're handling the space data center heat problems? Energy harvesting in space is easy, while getting rid of heat is quite hard.

by zulux 3mo ago

Any idea how they're handling the space data center heat problems? Energy harvesting in space is easy, while getting rid of heat is quite hard.

piloto_ciego 3mo ago

Scott Manley had a really good video on it that more or less confirmed my "bar napkin approximations" when people first started talking about this stuff. I think it's this one? https://youtu.be/DCto6UkBJoI?si=opcoaBEqKmw4LCrF https://youtu.be/DCto6UkBJoI?si=opcoaBEqKmw4LCrF Anyway, I think sun synchronous orbit (basically slightly off from polar) and pointing the solar panels toward the sun and the radiator away... It's not as far fetched as people would have you believe.

JumpCrisscross 3mo ago
Yup. It’s weirdly closer than I would have thought before sitting down with pen and paper.
jordanb 3mo ago
Did we watch the same video? My takeaway from Manley is that it is something that can pencil out as a physics problem with a very carefully selected orbit But as an engineering problem it's quite another thing. For example he talks about how far you have to pump the ammonia, and then you either need a very large amount of ammonia, or you have to pump a smaller amount of ammonia very fast through kilometers of plumbing. Of course there are highly theoretical ideas like making an atmosphere of coolant around the satellite or something but he left that stuff on the sci fi category.
gamblor956 3mo ago
That's like saying that controlled fusion is a napkin problem. The issue has always been the engineering. Programmers tend to be extremely bad at estimating he complexity of real world problems.

maxboone 3mo ago

https://www.nextbigfuture.com/2026/02/spacex-has-the-best-radiative-cooling-solution-and-it-is-working.html https://www.nextbigfuture.com/2026/02/spacex-has-the-best-ra...

zulux 3mo ago
Fair point, and thanks. Certainly true that Starlink satellites are already successfully dealing with the problem and use a lot of energy. I think the trouble with my thinking was that I was locked into the Shuttle/STS numbers of waste heat transfer.