Hacker News Reader

Comment by blauditore | Hacker News Reader

Why? I know people always mention billions of investment "until we have fusion". But those things cannot be reliably predicted, guesses are all pretty wild.

In terms of human working time, £200M roughly pays for 1000 person years of work, so it's certainly more than "a drop of water". Yes, I know there's more than salaries and research tools and material are expensive in this case. But still, imagine 50 people just working on theory and cheap experiments for 20 years; they could achieve a lot.

View on HN

200 million Euro would cover the raw material cost of the HTS material of a medium-large (3m major radius) scale reactor.

You still need to work the HTS material into magnets, a vacuum vessel, turbo and mechanical pumps, lots of fittings and mechanical engineering work, gyrotrons and their high voltage supplies and all the engineering that goes with (including wave guides), neutral beam injection, two dozen diagnostic systems, coil power supply system, and all of the work that goes with cryogenics.

Edit: If this would be for a medium scale science machine then you would need divertor and first wall materials, as well as a decent cooling system if you want long pulse operation. If this is a power plant then you can ditch most diagnostics, make everything twice as large, add a breeding layer between the coil cryogenics and the vacuum vessel, add a tritium separation facility ($$$), and add a big ass heat exchange/steam turbine system. ITER is in between these two and is the first time it's ever been done, with all of the diagnostics and full D+T operation caked into the design. It's not surprising it costs 20 billion Euro and even then if it was a power plant it wouldn't be priced out of reasonability (~twice current electric costs). At what point do we say we just need to pay more for our energy and kick fossil fuels?

As an aside, one of the design goals of the Commonwealth Fusion/MIT team was to lower the construction cost of HTS magnets to 100% of the material cost. So something like $100M for the superconducting tapes and another $100M to build them into magnets. Current superconducting magnets cost 10 times the material cost to construct.

Basically this level of funding gives Tokamak Energy the ability to order purchases of HTS tape in bulk and start ramping up to building all their full-scale magnets as they get deliveries over the next 3-5 years.

Edit: I find nothing in the announcement that says this is going towards magnets or HTS material, but this would be a the scale of advance funding for 3 meter radius magnet development.

Further Edit: The numbers are for SPARC which is about a 1.65 meter radius. 3 meters is the smallest size for a power plant, so on the order of $1.2 billion for the magnets assuming no significant decrease in material cost.

Building a conventional fission plant can easily cost $10B: https://thebulletin.org/2019/06/why-nuclear-power-plants-cos...

$200M is laughable - as is the 20 year plan. I recently reread Summa Technologicae by Stanislaw Lem. Written in 1961, it also guessed that fusion will be there 20 years in the future - meaning 1981 :)

>Building a conventional fission plant can easily cost $10B. $200M is laughable

The budget for ITER is 20b. The UK is only pitching in 200m. Other countries are pitching in the rest. The USA already pitched in $1b for example. The article only mentioned the UK because it was written for UK readers.

Hopefully this means that we’ll never achieve nuclear fusion in a practical way. As far as I can understand this will practically create energy at close-to-nothing prices, which while it will certainly solve some of our today’s issues (burning fossils is quite bad) at the same time it will also create a lot other bigger issues. For example, if you make it energy-free (i.e. price-free) to cut trees and build highways wherever you want I’m pretty sure the Amazon rain-forest will disappear a lot faster compared to what’s happening now.

We also need to capture CO2, not just stop producing it.

Nuclear fusion has a chance of being the solution to that problem.

No, fusion will not create energy at "close to nothing" prices. It promises to be extremely costly, actually.

Great, then it won’t happen (probably I had mistaken it with something else). Everything that makes energy less expensive and/or more easily available creates lots and lots of bad externalities. It’s time we stop believing that technology and technologic advancement will take us out of this mess, it is technology that threw us under the bus in the first place.

AFAIK the basic science has advanced enough that cheap experiments are not the best way forward. We basically know that designs like ITER work if you build them sufficiently large. There are still some engineering hurdles to be taken, and there might be better designs that can be built cheaper (e.g. Stellarators?), but we're at the stage where we need to build big machines if we want to progress rapidly.

The limiting factor for ITER was the strength of magnetic fields that could be attained from the superconductors available at the time. The promise for ARC/SPARC at MIT is that the field strengths available have recently doubled, and performance goes something like the 4th power of field strength, so a smaller reactor can reach the conditions that only a larger one could have reached when ITER was designed.

These smaller reactors should be cheaper and faster to build.

The limiting factor for high temperature superconductor magnets will likely be the mechanical strength of the structural steel holding the magnets together, not the limitations of the HTS. This is not a bad thing, just that the HTS are so much better than traditional superconductors at handling high magnetic field strength that you now have to optimize for the steel enclosures.

This is actually refreshingly exciting, coming from the usual "interstellar travel is slow and depressingly hard, space elevators are iffy at best, we probably aren't smart enough to build programmable nanotech" we get from most news updates on future tech.

Are suitcase-sized fusion reactors still on the table, or does this just mean smaller-than-a-football-stadium fusactors are probably possible?

You still need a one meter thick blanket of lithium to capture the neutrons from the D-T reaction (and to generate more Tritium) so you are looking at modular 500MW reactors, although locating multiple reactors on the same site makes sense for fueling and maintenance purposes.

Reply on news.ycombinator.com