Hacker News Reader

Comment by Retric | Hacker News Reader

Those are very different projects with very different goals. ITER is designed conservatively to produce 10x fusion energy output vs electrical energy input. As in you out 1 MW of electricity in and get 1 + 10 = 11MW of heat output. While not designed to be an economic fusion system it would unequivocally demonstrate a working design and a provide a test bed for experiments relating to reactor materials and full scale operations.

This however is aimed to explore novel geometries which may aid in future designs. However, the net efficiency at this scale would be vastly lower. It’s still a useful project, but it’s just one of many possible designs.

View on HN

It's not 10-11MW heat out for 1MW electricity in. It's "500 MW plasma from 50 MW of injected thermal power" which would need 300 MW electrical input. After conversion back to electricity, it might give as much electricity out as is put in!

See these links for more: Question asked in the EU parliament: https://www.europarl.europa.eu/doceo/document/E-8-2018-00081...

https://news.newenergytimes.net/2020/11/03/open-letter-to-it...

Useful clarification, but I think ITER’s website makes it unambiguous.

For 50 MW of injected heating power it will produce 500 MW of fusion power for long pulses of 400 to 600 seconds. https://www.iter.org/sci/Goals. To be more clear, injected heating power excludes the energy required to get to a steady state, as otherwise pulse length needs to be factored in.

That said ITER is running a lot of equipment unnecessary for a commercial power plant making engendering Q factors misleading. For example a power plant would not need to collect, process, and store the vast amount of scientific data that’s the point of the experiment. It’s really better if their not turning the rooms lights off trying to hit some arbitrary target.

PS: It was posted ~3AM and I am still out of it. So, I am kind of surprised how coherent that was even though I mess up conveying some heat was from electricity where the other heat was from fusion.

Why would it not be economic?

After the first reactor, do we expect building a replica, without all the R&D costs would be significantly more expensive than a standard nuclear reactor?

To me, 10x the energy sounds really good.

10x energy doesnt mean 10x electricity. How one captures that energy and converts it effectively remains an open question. Can all those particles be turned into heat? Will those particles degrade the reactor so fast that it is no longer economically viable?

ITER seems to be addressing the "capture the energy" part?

https://www.iter.org/sci/MakingitWork

"The neutrons will be absorbed by the surrounding walls of the tokamak, where their kinetic energy will be transferred to the walls as heat. In ITER, this heat will be captured by cooling water circulating in the vessel walls and eventually dispersed through cooling towers. In the type of fusion power plant envisaged for the second half of this century, the heat will be used to produce steam and—by way of turbines and alternators—electricity."

And without breeding blankets, yes you can capture the heat, but you rapidly run out of tritium. ITER will only be operate on DT for a few weeks, total over its lifetime, before the available tritium supply is exhausted (and its materials reach their neutron limits).

A "real" reactor will need to have tritium breeding blankets to regenerate the tritium. Otherwise, the cost of tritium alone, if made in fission reactors, for a 1 GW(e) DT fusion reactor would be $15B/year.

Isn't testing that approach of breeding tritium explicitly part of what ITER is trying to do with its TBM programme?

ITER has a few places where test blanket modules can be installed. ITER will not be able to simulate the neutronic environment of a full reactor (and so will not be able to confirm adequate breeding), nor subject the modules to the cumulative neutron loading they will have to withstand.

For this reason, a Fusion Nuclear Science Facility has long been proposed as something that will be needed before DEMO is built. This could be a small, Q < 1 DT fusion device, as the goal would not be plasma physics related. It itself will need to breed its own tritium to operate long enough, so there's a circular dependency in the development.

There are many such technology advances needed for fusion power to work. This is why spending $22 billion on ITER is a very inefficient use of money for research. Maybe as a diplomatic effort of having all the world's major powers [] (China, the European Union, India, Japan, Korea, Russia and the United States) work together it is worth it. Sort of like the space station. Big symbol and cost but little scientific value per dollar.

Unfortunately the budget for ITER comes out of the US science budget instead of the State Department. Most US fusion funding only goes to ITER at this point.

Fusion blankets are not just needed for fusion to work, they are needed for the next fusion reactor experiment (the one that has a burning plasma) to run. This is not a technology that we'll need some day for commercialization; it's a requirement for DT fusion to progress at all.

That's not capturing in a utility sense, it's merely preventing system failure from overheat (edit: autocorrect was turning this into "merrily preventing" which was kind of fitting as well). The result is a high volume, low delta-T coolant cycle and those are notoriously difficult (read: impossible) to capture meaningful amounts of electricity from, it will actually consume a lot. That quote is just a slightly more elaborate way of saying: no, we'll definitely not do any utilizeable capture.

How is what they just said any different than a standard nuclear reactor that just heats up water to produce steam into electricity?

Because ITER is not designed to be running for more than a few minutes at a time. Its successor, DEMO, will be designed for continuous power output.

https://en.wikipedia.org/wiki/DEMOnstration_Power_Station

Reply on news.ycombinator.com