4 ms·

Is fusion success in sight?

13 points by edw519 17y ago

CWuestefeld 17y ago
Summary: researchers are hopeful of achieving break-even in a controlled fusion reaction this year, based on extrapolations of small-scale experiments. But it's still a long way from an actually commercially-viable process, and there are other approaches that may be more successful -- or none at all.
- stcredzero 17y ago
  The Inertial Containment folks are going to go for exciting a hohlraum with their 192 lasers. I don't think blasting a precisely machined gold-plated metal cylinder every second is going to be economically viable.
  - mattheww 17y ago
    If they can reach create a controlled fusion reaction, a commercial reactor would be designed to continually feed fuel into it to continue the reaction. Reaching "break-even" is when the lasers don't have to be fired for the reaction to continue.
    - CWuestefeld 17y ago
      Reaching "break-even" is when the lasers don't have to be fired for the reaction to continue No, that's a "self-sustaining" reaction. Break-even means that the amount of energy released in the reaction is equal to the energy necessary to fuel the reaction.
- jerf 17y ago
  I can't even imagine how to make this commercially viable. Even if they reach breakeven, I can't imagine a process that would take less than a minute to reset the reactor for the next run, given the incredible precision necessary with the target, and I suspect my "minute" is grotesquely optimistic. I guess what this goes to show is that "breakeven" itself isn't even all that interesting. "Commercially viable" is what we are looking for. Here's hoping Polywell works out. Vaguely promising signs seem to be coming from the project but it's hard to see much detail through the Navy classification.
NathanKP 17y ago
I thought one of the statements in the article was especially interesting: "NIF isn't truly about energy," Seife writes in his book. "It is not about keeping our stockpile safe, at least not directly. It is about keeping the United States' weapons community going in the absence of nuclear tests." That's obviously a pretty pessimistic point of view, but when I consider the details of the test, including the immense amount of energy required to fire the lasers ten times a second, as would be required to start it according to the article, I don't see how it would be possible to get that much power back out, at least not using the traditional water to steam technique.
- electromagnetic 17y ago
  Essentially if they can make the equipment small enough, there could be serious potential for a pure-fusion bomb, which could skirt the test ban treaty as it doesn't fall under the terminology of a nuclear weapon (fissile or fissile and fusion; pure fusion isn't in the CTBT terminology, IIRC). Although the Ares 1 under development is currently capable of putting 25,000KG into LEO, and the Ares V is expected to be capable of up to 188,000KG. So it's not like the equipment has to be exceptionally small if it's capable of producing pure-fusion, it just has to be slightly smaller than the ISS!
jroes 17y ago
Aren't encryption algorithms measured by the amount of computing power it would take to crack them? If so, given a limitless power source, we could be in for some real privacy trouble, correct?
- lolcraft 17y ago
  Power in the street = muscle Power in politics = money Power in physics = watts Power in computer science = FLOPS Different meanings for different contexts.
- jerf 17y ago
  I do not know if you're being funny or not, because there is indeed a study of the minimum raw energy requirements for a given computation in our universe[1], so it is technically possible to rate the amount of computation something requires in terms of joules or your choice of energy unit. So I don't know if you know of this and are confused, or if you're trying to make a bad joke/pun on "power" (seems to be lolcraft's theory). Since there is a serious angle to your post, deliberate or otherwise, a serious answer: Because encryption difficulties tend to scale exponentially as you add bits, it is relatively easy to make encryption that brute forcing would require more energy than exists in the known universe. As big as the Universe is, exponential is "bigger". That's why even now the real problem is algorithms that have something far more clever than "brute force" available; we could already guarantee that even a post-Singularity adversary can't brute force our encryption (assuming they don't get new physics to play with), what we can't guarantee is that such an intelligence couldn't find a far more practical attack. [1]: http://en.wikipedia.org/wiki/Limits_to_computation http://en.wikipedia.org/wiki/Limits_to_computation
  - snissn 17y ago
    maybe he read the singularity is near too many times while stoned?

13 points by edw519 17y ago

Summary: researchers are hopeful of achieving break-even in a controlled fusion reaction this year, based on extrapolations of small-scale experiments. But it's still a long way from an actually commercially-viable process, and there are other approaches that may be more successful -- or none at all.

stcredzero 17y ago
The Inertial Containment folks are going to go for exciting a hohlraum with their 192 lasers. I don't think blasting a precisely machined gold-plated metal cylinder every second is going to be economically viable.
- mattheww 17y ago
  If they can reach create a controlled fusion reaction, a commercial reactor would be designed to continually feed fuel into it to continue the reaction. Reaching "break-even" is when the lasers don't have to be fired for the reaction to continue.
  - CWuestefeld 17y ago
    Reaching "break-even" is when the lasers don't have to be fired for the reaction to continue No, that's a "self-sustaining" reaction. Break-even means that the amount of energy released in the reaction is equal to the energy necessary to fuel the reaction.
jerf 17y ago
I can't even imagine how to make this commercially viable. Even if they reach breakeven, I can't imagine a process that would take less than a minute to reset the reactor for the next run, given the incredible precision necessary with the target, and I suspect my "minute" is grotesquely optimistic. I guess what this goes to show is that "breakeven" itself isn't even all that interesting. "Commercially viable" is what we are looking for. Here's hoping Polywell works out. Vaguely promising signs seem to be coming from the project but it's hard to see much detail through the Navy classification.

NathanKP 17y ago

I thought one of the statements in the article was especially interesting: "NIF isn't truly about energy," Seife writes in his book. "It is not about keeping our stockpile safe, at least not directly. It is about keeping the United States' weapons community going in the absence of nuclear tests." That's obviously a pretty pessimistic point of view, but when I consider the details of the test, including the immense amount of energy required to fire the lasers ten times a second, as would be required to start it according to the article, I don't see how it would be possible to get that much power back out, at least not using the traditional water to steam technique.

electromagnetic 17y ago
Essentially if they can make the equipment small enough, there could be serious potential for a pure-fusion bomb, which could skirt the test ban treaty as it doesn't fall under the terminology of a nuclear weapon (fissile or fissile and fusion; pure fusion isn't in the CTBT terminology, IIRC). Although the Ares 1 under development is currently capable of putting 25,000KG into LEO, and the Ares V is expected to be capable of up to 188,000KG. So it's not like the equipment has to be exceptionally small if it's capable of producing pure-fusion, it just has to be slightly smaller than the ISS!

jroes 17y ago

Aren't encryption algorithms measured by the amount of computing power it would take to crack them? If so, given a limitless power source, we could be in for some real privacy trouble, correct?

lolcraft 17y ago
Power in the street = muscle Power in politics = money Power in physics = watts Power in computer science = FLOPS Different meanings for different contexts.
jerf 17y ago
I do not know if you're being funny or not, because there is indeed a study of the minimum raw energy requirements for a given computation in our universe[1], so it is technically possible to rate the amount of computation something requires in terms of joules or your choice of energy unit. So I don't know if you know of this and are confused, or if you're trying to make a bad joke/pun on "power" (seems to be lolcraft's theory). Since there is a serious angle to your post, deliberate or otherwise, a serious answer: Because encryption difficulties tend to scale exponentially as you add bits, it is relatively easy to make encryption that brute forcing would require more energy than exists in the known universe. As big as the Universe is, exponential is "bigger". That's why even now the real problem is algorithms that have something far more clever than "brute force" available; we could already guarantee that even a post-Singularity adversary can't brute force our encryption (assuming they don't get new physics to play with), what we can't guarantee is that such an intelligence couldn't find a far more practical attack. [1]: http://en.wikipedia.org/wiki/Limits_to_computation http://en.wikipedia.org/wiki/Limits_to_computation
- snissn 17y ago
  maybe he read the singularity is near too many times while stoned?