[hackbinary]

300 MJ is about 83 KWh.

In the UK is 1 KWh is £0.34

So, this costs £28 in electricity to run this experiment. The experiment is a momentary thing.

Clearly there is now some work to, but now this is becoming an engineering problem of how to extend, sustain, and scale this process.

[tuatoru]

To have a hope of supplying grid power, they need to scale up the energy gain by four orders of magnitude and reactor run time by 12 orders of magnitude.

Those are just two of the engineering problems. It'll be a while, and I doubt it will ever compete with solar, wind, and storage.

[hackbinary]

Remember that the human genome project was started in in 1990, completed in 2003. In 2003 it took a year to sequence 80% of the human genome, now it takes a day to do 98%. In 1990 they had the understanding on how to sequence DNA, but scaling that across whole human genome seemed like monumental task.

I'm guessing in a decade we'll have a viable early stage industrial process, and in 2 decades we have commissioned fusion reactors.

If we don't kill the planet with nuclear war or the climate crisis.

[eru]

> [...], and I doubt it will ever compete with solar, wind, and storage.

Maybe not on earth, but there are applications in deep space.

[rbanffy]

> Maybe not on earth, but there are applications in deep space.

Depends on how long the interstellar craft is supposed to travel. If it's under 100 years, fission should be able to do the trick of keeping the craft warm and the lights on for the sealed ecosystem to function during the decades of coasting between stars.

Fusion rockets would be more convenient than fission ones because you can store the hydrogen you need in the form of water and water also acts as a great radiation shield while in deep space. Then, to brake, you use your radiation shield as reaction mass for fission or fusion rockets.

If we are talking about much more than that, fusion is probably a better answer as fission fuel will half-life itself into paperweights over a grand transgalactic tour.

[eru]

> Depends on how long the interstellar craft is supposed to travel.

I didn't have (only) travel in mind. I was thinking of living between the stars.

[rbanffy]

You'll need a constant power supply for the entirety of the trip. After doing the math, it turns out, fission seems quite viable - all usual fuels, in storage, have half-lives of more than 10,000 years.

So, if you have enough fissiles for keeping the closed ecosystem happy for the duration of the flight, you can go quite far.

The ship/colony will need to enter orbit around a star and drop by a rocky planet at some point, to gather more fissiles and reaction mass (and other materials needed for fixes and upgrades), so it wouldn't be able to stay indefinitely in deep space. If it's fusion-driven, a gas giant may be a good option for both fuel and reaction mass, and icy moons may work well for replacing water.

[eru]

I'm not talking about travel. I am talking about permanently living between the stars, and essentially living off hydrogen harvested from the interstellar medium.

[danw1979]

This is exactly why we need to keep plugging away at fusion - one day we’ll need it in space. It doesn’t have to be cheap, it just has to work.

[eru]

Of course, whether that's a good investment of resources right now is another question.

Even if we had no other goal than becoming an intergalactic species as soon as possible, we might still benefit from working on other things first.

[bratbag]

You are operating under the rather naive assumption that the people and resources used in fusion research are fungible with the people and resources used in other things.

When you have a bunch of people who know how to build nuclear bombs sitting around with nothing to do, you damn well keep them busy before another country finds them a job.

[eru]

For further discussion, we'd need to be careful whose perspective we are taking and what timescales we are talking about.

You seem to be taking the perspective of individual countries? And not eg humanity.

And timescales of perhaps decades?

On longer timescales: people don't get born knowing how to build nuclear bombs. They are trained up.

[doctorwho42]

Having worked in the field for 6 years, the estimated cost per shot at NIF is roughly $1MM. The estimated cost for a day of shots at OMEGA is $250k-$300k.

The cost per target varies a lot due to the precise manufacturing tolerances and the methods to get them. For example, the sphere with the fuel in it is made by dropping liquid glass from a drop tower. And then metrology is done on hundreds and hundreds of glass spheres.

So though the electricity might cost that, we are talking about a building in which just the lasers and their optical paths take up 3 foot ball fields of advanced warehouse space. And the target chamber is at ultra high vacuum, which is 10 meters in diameter. There are also countless diagnostics, computers, and other electronics, the lights for all the facility, and the number of people required to run it so this delicate experiment goes off without a hitch.

Honestly, it's almost not worth talking about as a power source anytime soon. Even if Q > 2 on NIF there are countless engineering problems that would have to be overcome (and haven't really been thought too hard on in the ICF field) to get a power reactor out of this tech.

My two cents, look towards MIT and CFS for news on their SPARC tokamak and plans for ARC tokamak. Based on some data I have seen, SPARC should hit Q>1 pretty easily. With some estimates of reaching Q> 3 to 9. And before you scoff at it, this reactor design is using magnetic tech that has proven it can withstand and produce a 20T magnetic field! In MCF, field strength and heating are the two key metrics. To put this into perspective, the massive tokamak being built in Europe has a MAX possible field strength of 13T, assuming it's run to the edge of it's theoretical design limitations. The SPARC one hasn't even been run to it's design limitations, most likely due to the mechanical stresses a 20T field produces in a 3-4 meter D coil.

[moloch-hai]

Plus $10M for the target bead.

[philipkglass]

Where did you see this estimate for hohlraum cost?

The NIF fired 368 shots in 2021:

https://lasers.llnl.gov/for-users/nif-target-shot-metrics

At $10 million per target that would cost $3.7 billion. The annual LLNL budget (which includes NIF) is only $2.8 billion:

https://www.llnl.gov/doing-business/economic-impact

As of 2004, the targets were reported to cost $2500 each:

https://www.osti.gov/servlets/purl/828518

[moloch-hai]

Got it from a comment on here.

The 2004 targets didn't work. Neither did the 368 shots in 2021. Maybe ones that work cost more?

Even $2500 for 1 kWh is rather steep.