6 ms·

the only answer is nuclear, simply because of energy density required for space applications. the creators of the original solid-core Krusty/Kilopower reactor

the only answer is nuclear, simply because of energy density required for space applications. the creators of the original solid-core Krusty/Kilopower reactor are trying to commercialize the tech, I hope they have enough funding (@sama, I hope this is on your radar) https://www.spacenukes.com/ https://www.spacenukes.com/ This is the first fundamentally new reactor actually built and tested by NASA in more than 50 years (vs. countless paper designs) https://www.nasa.gov/mission_pages/tdm/fission-surface-power/index.html https://www.nasa.gov/mission_pages/tdm/fission-surface-power... EDIT: I highly recommend watching the presentation by David Poston "Space Nuclear Power for Mars" https://www.youtube.com/watch?v=luQfEYs2L0w https://www.youtube.com/watch?v=luQfEYs2L0w

baybal2 5y ago
I don't see battery storage as a problem for the first outpost, but yes, kilopower reactors were made exactly for this. Another option for early missions is just to limit them to 14 days. Heating of the base can be achieved by preheating the regolith deep beneath the surface. With regolith density we know, it should be well feasible. Thermal insulation can be made very efficient in space because there is no atmosphere. I see a nuclear/rtg base power for life support, and solar powered regolith smelting following the 14 day cycle.
- Robotbeat 5y ago
  That’s pretty reasonable. Particularly since nuclear (fission or radioisotope) produces a lot more heat than electricity (thermal to electrical conversion efficiency is just 23% for Kilopower and 3-9% for RTGs), so using it for base heating during the night is a much better argument. And even some solar powered rovers like Spirit and Opportunity used small radioisotope heat sources to stay warm.
Robotbeat 5y ago
“The answer is nuclear” is repeated often but often is an innumerate answer. Kilopower for most uses in the inner solar system performs considerably worse than solar. The power to weight ratio of Kilopower is about 6-7W/kg, no better than old time RTGs. UltraFlex solar panels do about 150W/kg near 1AU. Nuclear in the lunar case may be useful for base power due to the long lunar night which makes storage pretty heavy, but nuclear is actually usually WORSE performance than solar, producing much less power for the same mass. It’s only in the outer solar system around Jupiter or in niche cases like lunar night where nuclear has a solid lead on solar. Otherwise it’s similar or worse, besides being a lot more expensive. ~$100 million for 10kWe Kilopower.
- andyxor 5y ago
  i think you are missing the point of Kilopower, it was to build a working micro reactor prototype on realistic timeframe and actually ship it, they have designed the extension of this to MW scale where energy density is order of magnitude better. Nuclear scales better than any other source of energy. How many sq feet of solar panels do you need for 1, 10, 100, 1000, 10000 MWe, now translate it to lbs and $$$ cost (including shipping to the moon or mars base), also consider dust and lunar night conditions affecting solar use. if you look at scaling it vs. weight and cost (including maintenance and replacement panels over 20-30 years), solar is no competitor to nuclear.
  - Robotbeat 5y ago
    No, the larger nuclear power plants (50kWe) NASA has developed to reasonable TRL are still around the same power to weight ratio. Nuclear, in addition to everything else like the dynamo and the reactor itself, requires heavy radiators. Also, Kilopower has a SHORTER lifespan of about 15 years whereas solar can last 30-50 years (solar degradation is much slower than is often claimed, at least when we’ll engineered and in the absence of high humidity) as there’s no moving parts and nothing to refuel. Remember, terrestrial nuclear reactors are regularly refueled (annually?) and maintained. The traditional argument has always been that nuclear scales better than solar, but usually such comparisons are looking at sandbagged, outdated solar panels and very low-TRL nuclear designs relying on super high rejection temperatures (ie unrealistic) or shorter lifespans. I’m not anti-nuclear by any stretch, and I fully support increased funding and deployment of nuclear power, but I think there has been a lot of hand waving by some nuclear advocates. If you look at actually achievable, near term nuclear designs, none of them perform anywhere near as good as solar does in orbit near 1AU, and it’s not even close. It’s surface applications or outer solar system where nuclear has a chance against solar. Outside of those cases, solar is much more powerful for a given weight (meaning cheaper to transport) plus being far cheaper to build and not requiring special launch vehicle risk mitigations (currently, the only nuclear-rated launch vehicle is Atlas V… and every nuclear launch must be approved by the executive branch). (A human Mars base may benefit from a mix of power sources for resiliency purposes.)
- slfnflctd 5y ago
shkkmo 5y ago
On the moon, Nuclear has basically the opposite problem of solar: Long, hot days mean that nuclear power generation is hard to cool during the lunar day. This means that nuclear for the moon is basically directly competing with batteries and other storage solutions in terms of weight/risk/difficulty.

baybal2 5y ago

I don't see battery storage as a problem for the first outpost, but yes, kilopower reactors were made exactly for this. Another option for early missions is just to limit them to 14 days. Heating of the base can be achieved by preheating the regolith deep beneath the surface. With regolith density we know, it should be well feasible. Thermal insulation can be made very efficient in space because there is no atmosphere. I see a nuclear/rtg base power for life support, and solar powered regolith smelting following the 14 day cycle.

Robotbeat 5y ago
That’s pretty reasonable. Particularly since nuclear (fission or radioisotope) produces a lot more heat than electricity (thermal to electrical conversion efficiency is just 23% for Kilopower and 3-9% for RTGs), so using it for base heating during the night is a much better argument. And even some solar powered rovers like Spirit and Opportunity used small radioisotope heat sources to stay warm.

Robotbeat 5y ago

“The answer is nuclear” is repeated often but often is an innumerate answer. Kilopower for most uses in the inner solar system performs considerably worse than solar. The power to weight ratio of Kilopower is about 6-7W/kg, no better than old time RTGs. UltraFlex solar panels do about 150W/kg near 1AU. Nuclear in the lunar case may be useful for base power due to the long lunar night which makes storage pretty heavy, but nuclear is actually usually WORSE performance than solar, producing much less power for the same mass. It’s only in the outer solar system around Jupiter or in niche cases like lunar night where nuclear has a solid lead on solar. Otherwise it’s similar or worse, besides being a lot more expensive. ~$100 million for 10kWe Kilopower.

andyxor 5y ago
i think you are missing the point of Kilopower, it was to build a working micro reactor prototype on realistic timeframe and actually ship it, they have designed the extension of this to MW scale where energy density is order of magnitude better. Nuclear scales better than any other source of energy. How many sq feet of solar panels do you need for 1, 10, 100, 1000, 10000 MWe, now translate it to lbs and $$$ cost (including shipping to the moon or mars base), also consider dust and lunar night conditions affecting solar use. if you look at scaling it vs. weight and cost (including maintenance and replacement panels over 20-30 years), solar is no competitor to nuclear.
- Robotbeat 5y ago
  No, the larger nuclear power plants (50kWe) NASA has developed to reasonable TRL are still around the same power to weight ratio. Nuclear, in addition to everything else like the dynamo and the reactor itself, requires heavy radiators. Also, Kilopower has a SHORTER lifespan of about 15 years whereas solar can last 30-50 years (solar degradation is much slower than is often claimed, at least when we’ll engineered and in the absence of high humidity) as there’s no moving parts and nothing to refuel. Remember, terrestrial nuclear reactors are regularly refueled (annually?) and maintained. The traditional argument has always been that nuclear scales better than solar, but usually such comparisons are looking at sandbagged, outdated solar panels and very low-TRL nuclear designs relying on super high rejection temperatures (ie unrealistic) or shorter lifespans. I’m not anti-nuclear by any stretch, and I fully support increased funding and deployment of nuclear power, but I think there has been a lot of hand waving by some nuclear advocates. If you look at actually achievable, near term nuclear designs, none of them perform anywhere near as good as solar does in orbit near 1AU, and it’s not even close. It’s surface applications or outer solar system where nuclear has a chance against solar. Outside of those cases, solar is much more powerful for a given weight (meaning cheaper to transport) plus being far cheaper to build and not requiring special launch vehicle risk mitigations (currently, the only nuclear-rated launch vehicle is Atlas V… and every nuclear launch must be approved by the executive branch). (A human Mars base may benefit from a mix of power sources for resiliency purposes.)
slfnflctd 5y ago

shkkmo 5y ago

On the moon, Nuclear has basically the opposite problem of solar: Long, hot days mean that nuclear power generation is hard to cool during the lunar day. This means that nuclear for the moon is basically directly competing with batteries and other storage solutions in terms of weight/risk/difficulty.