Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Revamped German stellarator should run longer, hotter and compete with tokamaks

science.org

Post reply on HN
Print view View original

61–70 of 157 posts

Re: Revamped German stellarator should run longer, hotter and compete with tokamaks

#61

Post by kortex »

onlyrealcuzzo wrote: post #17
sidkshatriya wrote: post #12

Earlier quoted context omitted.

From what I've understood from previous discussions (don't know much about plasma physics though): It is a multistep challenge. Step 1: Figure out how to get a self-sustaining fusion reaction. Step 2: figure out how to extract the energy in a useful, safe, non-destructive (to the reactor) way. Doing Step 1 alone is very difficult so the've postponed Step 2 to the future. I may be wrong though.

I would imagine since, combined, we're spending tens of billions of dollars on Fusion research - someone must have some ideas for step 2, right?!

We have some ideas, of course. The rough idea is: surround the reactor with a lithium blanket. This gets bombarded with neutrons, breeding fuel and heating the lithium, which transfers its heat to coolant to spin turbines. The devil is in all the engineering details, like: how do we make alloys to deal with neutron bombardment/hydrogen embrittlement, what's the optimal geometry to capture flux while being serviceable, etc.

Much of that work can and is researched in parallel, but there will be inevitable integration hell that has a burning fusion reactor core as a prerequisite.

Re: Revamped German stellarator should run longer, hotter and compete with tokamaks

#62

Post by dmichulke »

pfdietz wrote: post #53
GordonS wrote: post #29

Earlier quoted context omitted.

I agree. If we were able to eventually miniaturise fusion reactors, they would be incredible for space flight. Back on Earth, if they could be made safe enough, maybe we could even use them to run container ships?

DT fusion would be just ridiculous for space flight. Anything a DT reactor could do a fission reactor could do much better -- much smaller, much higher power density, much less complexity. And similarly for use in ships down here on Earth. In practice, synfuels would be better than either for ships on Earth.

Isn't the main reason for us not having fission rockets the fact that we don't want to have radioactive material explode a few kms above our head aka a dirty bomb?

Re: Revamped German stellarator should run longer, hotter and compete with tokamaks

#63

Post by amacbride »

phist_mcgee wrote: post #58
EamonnMR wrote: post #30

Earlier quoted context omitted.

Splark: a new phlinto framework built in blenk. And the thread is just Blenk fans arguing with Fobl fans.

As a fan of Fobl and their work on Spoogum, i'm not sure you would call it 'arguing' more sharing the obvious downsides of the dinglepop pipeline when combined with the grumpkin anti-pattern.

Meh. This was all solved decades ago with the turbo encabulator.

Re: Revamped German stellarator should run longer, hotter and compete with tokamaks

#64

Post by mjhay »

marcosdumay wrote: post #56
moogly wrote: post #41

Earlier quoted context omitted.

> We’re still a few years away from having them in actual power plants I admire your optimism. I'd say half a century, if things go well, but I'm not convinced fusion will ever be economically feasible.

> but I'm not convinced fusion will ever be economically feasible Not on Earth. But put a good enough fusion reactor on a rocket, and you can reach neighboring stars in 3 or 4 decades instead of the 15 one would expect for fission. (Of course, nobody is even sure reactors can get that good, but it does look possible.)

Fission fragment rockets could actually compete with fusion here. The reaction mass is the fission fragments themselves, which could get a specific impulse of around 10^6 s. The technology to do this is within relatively easy reach, at least by interstellar rocket standards.

https://en.wikipedia.org/wiki/Fission-fragment_rocket

Re: Revamped German stellarator should run longer, hotter and compete with tokamaks

#65

Post by pfdietz »

dmichulke wrote: post #62
pfdietz wrote: post #53

Earlier quoted context omitted.

DT fusion would be just ridiculous for space flight. Anything a DT reactor could do a fission reactor could do much better -- much smaller, much higher power density, much less complexity. And similarly for use in ships down here on Earth. In practice, synfuels would be better than either for ships on Earth.

Isn't the main reason for us not having fission rockets the fact that we don't want to have radioactive material explode a few kms above our head aka a dirty bomb?

A DT fusion reactor would have a thrust/weight ratio much less than 1, so it could not be used in a launch vehicle. It would be purely for use up in space. So, none of this "a few kms above our heads" nonsense.

Re: Revamped German stellarator should run longer, hotter and compete with tokamaks

#66

Post by mjhay »

Retric wrote: post #50
Maken wrote: post #49

Earlier quoted context omitted.

It feels like fusion energy is perpetually 50 years away.

Minimal investment means minimal progress. ITER was basically a Regan/Gorbachev project from 1985 that’s still not built yet. It’s JET’s the current largest device was completed in 1983. Real progress has been made, but the major projects have been extremely conservative by necessity. Early designs for ITER where for a larger device that would have actually produced electricity though not cheaply enough to be economi…

And ITER's design is already wildly outmoded now that high-temp superconductors exist (which they have for quite some time). Just a multinational pork project.

Re: Revamped German stellarator should run longer, hotter and compete with tokamaks

#67

Post by kingkawn »

pfdietz wrote: post #59
simonebrunozzi wrote: post #43

Earlier quoted context omitted.

I think that 50 years is a reasonable timeframe to give fusion the possibility of delivering results. But I agree with you, it's pretty much impossible that anything will come out of fusion in the coming 10-15 years, despite the publicity we keep hearing lately.

We've already given it more than 50 years.

Yes, but making long predictions allows them to pretend they are in pursuit of something

Re: Revamped German stellarator should run longer, hotter and compete with tokamaks

#68

Post by zasdffaa »

marcosdumay wrote: post #56
moogly wrote: post #41

Earlier quoted context omitted.

> We’re still a few years away from having them in actual power plants I admire your optimism. I'd say half a century, if things go well, but I'm not convinced fusion will ever be economically feasible.

> but I'm not convinced fusion will ever be economically feasible Not on Earth. But put a good enough fusion reactor on a rocket, and you can reach neighboring stars in 3 or 4 decades instead of the 15 one would expect for fission. (Of course, nobody is even sure reactors can get that good, but it does look possible.)

That is highly tangential to the issue of economic viability, and the energy crisis.

Re: Revamped German stellarator should run longer, hotter and compete with tokamaks

#69

Post by ethbr0 »

mjhay wrote: post #64
marcosdumay wrote: post #56

Earlier quoted context omitted.

> but I'm not convinced fusion will ever be economically feasible Not on Earth. But put a good enough fusion reactor on a rocket, and you can reach neighboring stars in 3 or 4 decades instead of the 15 one would expect for fission. (Of course, nobody is even sure reactors can get that good, but it does look possible.)

Fission fragment rockets could actually compete with fusion here. The reaction mass is the fission fragments themselves, which could get a specific impulse of around 10^6 s. The technology to do this is within relatively easy reach, at least by interstellar rocket standards. https://en.wikipedia.org/wiki/Fission-fragment_rocket

Neat links!

I think nuclear propulsion is going to be the biggest beneficiary of higher mass-to-orbit-for-a-reasonable-price advances.

There are a huge number of propulsion technologies that are physically possible but too heavy and/or dangerous for near-Earth use.

Cheaper lift (to bootstrap) + more ongoing destinations and transit work (to drive) + outside of Earth orbit (to alleviate safety concerns) = rapid progress

Re: Revamped German stellarator should run longer, hotter and compete with tokamaks

#70

Post by XorNot »

mjhay wrote: post #66
Retric wrote: post #50

Earlier quoted context omitted.

Minimal investment means minimal progress. ITER was basically a Regan/Gorbachev project from 1985 that’s still not built yet. It’s JET’s the current largest device was completed in 1983. Real progress has been made, but the major projects have been extremely conservative by necessity. Early designs for ITER where for a larger device that would have actually produced electricity though not cheaply enough to be economi…

And ITER's design is already wildly outmoded now that high-temp superconductors exist (which they have for quite some time). Just a multinational pork project.

ITER exists to prove that fusion plasma stability can be sustained to extract useful energy. How it does it is irrelevant since the big news is having a sufficiently large tokamak vaccum vessel with instrumentation to study.

HTSCs weren't usable when it was designed, and have only just become usable in the last 5 years or so but they are a fundamentally different material. You don't just drop them into a large, incredibly complex machine that depends on it's integrated magnetic containment system: you are functionally building a new device.

If you can ITER, then you don't get a refund on spent dollars. You get a loss. And then you get to start another 30 year project to maybe build a new vacuum vessel, which you have to do because you still haven't actually tested plasma stability.

"But but MIT skunkworks!"...yeah. It's still going along, and they haven't suddenly churned out a functioning reactor based on HTSCs because oh look, whatever the advantages they're a new material with different properties, manufacturing and handling behaviors all of which need to be developed, measured and inspected before you can use them effectively in a fusion device. If they look good then great: they can be used to make DEMO, the ITER-successor commercial powerplant prototype, cheaper and more powerful.

Post reply on HN