Liquid hot plasma
January 8, 2025 5:49 AM   Subscribe

It's been 8 years since we last checked in on ITER. How's it coming along?

July 2024:
The giant fusion reactor known as ITER will not turn on until 2034, 9 years later than currently scheduled, according to a new timeline the international organization announced this week. Energy producing fusion reactions—the goal of the project—won’t come until 2039, and only in short bursts, to satisfy safety concerns of the nuclear regulator in France, where ITER is under construction. ... With dozens of private fusion companies promising power production on shorter timescales, some say ITER may be obsolete by the time it turns on.
posted by Lemkin (34 comments total) 8 users marked this as a favorite
 
Great YouTube video and clearly developed with a high attention to detail and nice production values. ITER is a very impressive project though I do wonder if it's just too big and too slow to be successful (I very much hope I'm wrong here).

One thing I think of with these sorts of projects and the narratives around "they're always over budget and past deadlines" is the little dance that projects seeking government funding need to do.

(For the sake of this argument, set aside the current toxic political environment and funding for the sciences and the arts).

If NASA goes to Congress and says, "We need $10 billion to build a new space telescope and it will take 12 years" they're likely to get denied. It's too hard for congresspeople to go back and justify to constituents that sort of long-term commitment.

So instead NASA says, "We need $1 billion and it will take 4 years." And then, three years later they say, "Well, we've made great progress but we need an additional billion and two more years."

If you're a representative of senator at this point, you don't want to vote against something that was previously approved or you've now wasted that original billion you allocated. So you give them an extension and a longer leash. Repeat several times and you get close to the actual cost and timeline of the project.

It's not bureaucracy and government inefficiency and waste. It's playing a political game for funding. It's quite savvy, actually.

So, ITER going from a €6 billion budget to €20 billion doesn't surprise me in the least. I hope they are successful in the end and we get some valuable science out of it. Compared to the economic costs of fossil fuels and climate change, it's a negligible number.
posted by robot_jesus at 6:26 AM on January 8 [11 favorites]


[Insert 10 years away joke here.]

Beyond that, I cannot find the numbers, but I remember reading that this would be essentially commercially unviable. How much is this thing supposed to cost to operate vs solar/wind/natural gas? The promise of fusion was cheap, clean limitless energy. There's some very exciting science there, but I wonder if this will ever actually be a useful energy source.
posted by Hactar at 6:34 AM on January 8 [3 favorites]


The promise of fusion was cheap, clean limitless energy.

I’m sure we’d settle for a fallout-free substitute for nuclear weapons.
posted by Lemkin at 6:40 AM on January 8 [2 favorites]


If we really had cheap, clean, limitless energy, we would cook the Earth with our waste heat. Even if we only increased our energy usage by 2% a year using some magical perfectly clean, perfectly scalable energy source, we would hit the ocean-boiling point in just 400 years. That’s literal 100C boiling. Obviously we wouldn’t get that far before we all died. Air conditioning wouldn’t help either, it would just accelerate the problem via more energy use.

Note that this is strictly a thermodynamic argument. Unless we somehow break the laws of thermodynamics, that’s what 2% exponential energy growth results in. And 2% isn’t a number pulled from the air, it’s the approximate growth rate of human energy usage throughout the whole of the industrial age.
posted by notoriety public at 6:58 AM on January 8 [17 favorites]


It's not bureaucracy and government inefficiency and waste. It's playing a political game for funding. It's quite savvy, actually.

I'm not sure there's any daylight between those two things. Bureaucracy slows things down and spreads out accountability*. Requesting funding in smaller amounts at more frequent intervals...slows things down and spreads out accountability.

*Not a critique! Slowing things down to assure a problem is well thought out has its merits, as does spreading out accountability to assure the right decision-makers are being given the chance to make the decisions in their purview. I think "bureaucracy" being seen by the public as hand-in-hand with inefficiency and waste is one of the less acknowledged results of conservatives being very good at controlling language use to promote their agenda.
posted by solotoro at 7:09 AM on January 8 [2 favorites]


Unless we somehow break the laws of thermodynamics

Then that should be our next big project: Break the Laws of Thermodynamics

It’s already got a catchy acronym.
posted by Lemkin at 7:12 AM on January 8 [6 favorites]


I very much hope I'm wrong here

Serious question: why?

Fission power looks good compared to coal because you get so much more energy out per ton of fuel put in. Fusion power looks potentially great compared to coal for the same reason only more so. So I can see where the whole "unlimited" trope comes from.

But here's the thing: solar PV and wind and other technologies that likewise work by diverting existing energy flows all crap from a great height upon all the fuel-burning technologies on that very same basis, because you can get unlimited amounts of energy out of them for zero fuel put in.

Oh, say the critics, but we can't get unlimited amounts of energy out of a solar panel because the energy density of sunlight at the surface of the Earth is so low. To which I say: sure, 1kW/m2 input for less than half of each day doesn't seem too impressive on its face, but mass production is a thing and even at today's prices for PV panels and batteries and HVDC transmission lines and land, we can extract more power from that existing radiation than we'll ever be able to use without cooking ourselves. So what's the point of your big centralized expensive complex vulnerable brittle nukes with their complicated expensive vulnerable brittle environmentally damaging and generally dangerous fuel supply chains again? At which point they mostly resort to glowering and insults.

Obtaining a gigawatt of electrical power requires somebody to pay for a gigawatt's worth of generating plant. Nuclear fission plant lost the cost effectiveness race a decade ago even setting aside the inevitable decommissioning costs and as we can all clearly see, even getting fusion to happen at all, let alone at commercial scale, has already cost more than all the research ever put into fission combined. So why on God's green Earth are we still so all fired up to pursue this tech that we have already obsoleted? Obsoleted now, not even by the time ITER first turns on?

Near as I can tell, the motivation is literally about concentration of power; all non-medical nuclear research is really no more than a job creation scheme for the weapons industry.
posted by flabdablet at 7:18 AM on January 8 [19 favorites]


I cannot find the numbers, but I remember reading that this would be essentially commercially unviable.

I don't think there was ever any thinking that ITER would be economically viable in terms of its energy generation. Its more aimed at stimulating growth in the countries investing in it, while working on the many scientific and engineering barriers and maybe delivering some working tech that might inform subsequent generations of the technology which might be viable in the future.

As flabdablet says, renewables is now well ahead of both fusion and fission economically. ITER has taken so long that virtually all of the cost savings for both wind and solar (the two big RE success stories) have happened since it was mooted.
posted by biffa at 7:25 AM on January 8 [3 favorites]




If we really had cheap, clean, limitless energy, we would cook the Earth with our waste heat.

Any argument based on unbounded exponential extrapolation is suspect, to put it mildly

Almost nothing in the real world is well modeled in the long term by an exponential function. To put some actual data forward, per capita energy usage in the US was rising at about 3% per year until 2010. Since 2010 it has dropped by about 5% overall. If energy were cheap this number would certainly go up, but at some point you just run out of interesting things to do with more energy.
posted by Pemdas at 7:50 AM on January 8 [5 favorites]


If we really had cheap, clean, limitless energy, we would cook the Earth with our waste heat.

Not necessarily. Wind and solar both use energy that was already in the system, getting some work out of it doesn't impact the total energy in the system.
posted by biffa at 8:13 AM on January 8 [3 favorites]


> Wind and solar both use energy that was already in the system, getting some work out of it doesn't impact the total energy in the system.

Doesn't affect the total energy of the system, but turning kinetic energy of wind and tide into heat will raise the temperature, which was the complaint.

I took the liberty of reading the Wiki article about ITER, at least the first part that talks about goals. If ITER is a crashing win, it will use 300MW of power to push 50MW of heat into plasma, yielding 500MW of heat. Which will be vented, not used to create electricity. The "10x yield" part ignores that the creation of the 50MW of heat is only 15% efficient.

This is not a system that will revolutionize electrical power generation, even if it is an unqualified success within the current schedule, which as explicated elsethread is probably a political fiction.
posted by Aardvark Cheeselog at 8:54 AM on January 8 [3 favorites]


kinetic energy of wind

The kinetic energy of the wind comes from solar warming, so was already heat. The amount converted would be trivial even if all human demand was met from wind energy and then went up by a factor of ten. You might get some local wind effects but you will have to evidence anything more.
posted by biffa at 10:59 AM on January 8 [1 favorite]


Even DEMO, the 2050-ish successor project to ITER with the design goal of putting energy into the grid, is only projected to run for 5 minutes at a time and requires energy input that's about equal to what's generated.

There's nothing publicly disclosed that suggest any of the fusion startups can address any of the fundamental materials or engineering problems with fusion energy. The solutions to the challenges of a purely renewable grid (storage, distribution) are well understood and solvable with current technology.
posted by theclaw at 11:06 AM on January 8 [2 favorites]


We ought to be a bit careful with where we put solar PV; it's better not to put it in places where it reduces local albedo. But using it to cover up places whose albedo we've already fucked, like parking lots, is win-win-win. Especially once those lots fill up with electric vehicles that will happily slurp up all that PV electricity that genuinely does become too cheap to meter at peak insolation times, and take it home to run houses at night.
posted by flabdablet at 11:12 AM on January 8 [1 favorite]


The kinetic energy of the wind comes from solar warming, so was already heat.

Yeah, this was my intuitive reaction, but I'm not an engineer or physicist so I was wrestling a little. I feel like the analyses that suggest we'd cook ourselves kind of ignore that Earth radiates excess energy into space. I don't know how we could raise the global temperature that far using "cheap, clean, limitless" energy assuming that the Sun's output is basically constant. With cheap, limitless dirty energy, sure - we'd be messing with the planet's ability to radiate energy until things warm up enough to reach a new equilibrium (OK, not "we would be" but rather "we are"). But if we're just reorganizing the form of energy once it reaches us from the Sun, I don't understand where all this extra heat is coming from and why it can't be radiated into space.

I totally agree with the argument that exponential growth in any system has to have limits, which is why I was a little unsure about my gut reaction (and welcome any corrections/clarifications!), but it sounds like in this particular case, the limits aren't really thermodynamic.
posted by nickmark at 11:39 AM on January 8


> There's nothing publicly disclosed that suggest any of the fusion startups can address any of the fundamental materials or engineering problems with fusion energy.

Indeed. The "cleanness" of fusion is a bit of a bait-and-switch: it's not going to make barrels of waste from depleted fuel, but fusion reactions make lots of neutrons, so any fusion reactor that operates commercially is going to reach end-of-life having spent its entire service life being bombarded with high-level neutron flux. i.e. it will be high-level radioactive waste (much like the containment vessels of existing fission plants) and will have decommissioning costs to match.

That's after solving the materials problems for building a reactor that can turn heat from fusion plasma into electricity.
posted by Aardvark Cheeselog at 11:41 AM on January 8 [4 favorites]


> The kinetic energy of the wind comes from solar warming, so was already heat.

That got turned into kinetic energy by a process that involved cooling, yes. Generating wind power involves turning (part of) the kinetic energy back into heat, and using the power will make more heat. Same with tidal power.

But it is true, that harvesting so much wind power that the inefficiencies of its conversion raised global temperature... would plausibly cause catastrophic ecological change by itself, through disruption of processes that depend on air circulation patterns.
posted by Aardvark Cheeselog at 11:48 AM on January 8 [1 favorite]


Why would the inefficiency of conversion of wind energy impact global temperature? The wind not converted into energy stays as wind. The more inefficient ithe process is, the more stays as wind.
posted by biffa at 12:13 PM on January 8 [1 favorite]


Generating wind power involves turning (part of) the kinetic energy back into heat, and using the power will make more heat.

Which will get dumped directly or indirectly into the atmosphere and turn back into wind.

This ended up being A Thing about KSR's *Red Mars* -- Sax's plan to heat Mars with windmills could never work because all he was doing was turning wind into wind.
posted by GCU Sweet and Full of Grace at 12:17 PM on January 8 [2 favorites]


If we really had cheap, clean, limitless energy, we would cook the Earth with our waste heat.

Here is an article about how to reduce heat on the surface of the planet, and generate electricity in the same process:

Tackling Climate Change through Radiative Cooling

Is it perfect? Probably not—but I think it's a good option to investigate further.

I think the "if we got fusion working, we would cook the Earth" argument is shortsighted, because there are plenty of ways to use or dump heat, if we wanted to pursue those. And also, in talking about fusion, we usually assume that it would replace some other, much dirtier method of producing power which would very likely also produce heat.
posted by newdaddy at 12:32 PM on January 8 [1 favorite]


If we really had cheap, clean, limitless energy, we would cook the Earth with our waste heat.

If we really had cheap, clean, limitless energy, wouldn’t it make it more plausible to live somewhere beyond the Earth, or enable any number of other sci-fi ideas that might mitigate this exponential extrapolation? The bigger problem with “cheap, clean, limitless” would seem to be that none of those things are necessarily true.
posted by atoxyl at 9:32 PM on January 8 [1 favorite]


The amazing Omega Tau podcast had a 3 hour episode on ITER in 2014. (I thought it was more recent; maybe there's a followup that I'm not finding.) (Note that OT has some episodes in German and some in English.)
posted by neuron at 10:13 PM on January 8


I'm half asleep, but isn't it right that a fusion reactor works by generating heat that is used to turn water to steam which turns a turbine to generate electricity with some very small fraction of the heat energy becoming electricity and most of the fusion energy heating the environment?
posted by neuron at 10:20 PM on January 8


in talking about fusion, we usually assume that it would replace some other, much dirtier method of producing power

and those of us who cannot be persuaded to let go of the Gernsback Continuum then go on to ignore the glaringly obvious fact that we already have better, safer, more mature and cheaper technologies that can replace and are replacing those dirtier generators faster than fusion ever could, even given the best possible development timeline. Fission too, for that matter. Burning fuels is obsolete. Fusion has missed its window of opportunity and pursuing it further is simply not justifiable on the basis of any reasonable energy policy. And yet it is still being pursued at vast expense, which is the tell that the weapons trade is its primary driver.

if we're just reorganizing the form of energy once it reaches us from the Sun, I don't understand where all this extra heat is coming from and why it can't be radiated into space

Because in extracting work from it we've lowered its temperature, which shifts its black body radiation further down the spectrum and makes it easier for the atmosphere to intercept that radiation.

The effect is small, but then again, so is the effect of increasing CO2 concentration by a hundred ppm or two. The thing to keep in mind about the planetary radiative energy balance is that the quantities of energy involved are so mind-manglingly huge that even small changes to the way the balance works can be highly consequential.

Replacing all of Earth's fuel-burning technologies with solar PV, wind and battery storage while energy consumption stays roughly commensurate with today's would definitely move the radiative energy balance in the direction we need it to go, but there would be an effect on albedo and that effect should not be forgotten about.

The single most effective move is to reduce overall energy consumption via the relentless pursuit of energy efficiency, Jevons notwithstanding. Predicating the future of civilization on endless growth in energy consumption (or anything else, really) is just plain stupid, as everybody except the c*nts in charge now understands.
posted by flabdablet at 12:13 AM on January 9 [2 favorites]


I feel like the analyses that suggest we'd cook ourselves kind of ignore that Earth radiates excess energy into space.

We've already started to cook ourselves from making a tiny change to the Earth's ability to do exactly that.

Any feels that any of us have about these things are highly likely to mislead us because, as I mentioned above, the sheer size of the energy flows involved in this planet's radiative energy balance is just so far beyond most people's intuitive comprehension.

Imagine the most powerful hurricane you've ever seen reported. The most powerful tornado. The biggest flood. The biggest duststorm. The biggest tsunami. The worst wildfire. Think about the massive size of any of these things compared to the powers and capabilities of your puny human organism and how terrifying and awe-inspiring that makes those phenomena and what a reality check they offer against human hubris.

Now imagine that you are that hurricane or that tornado or that flood or duststorm or tsunami. You're hot shit now, huh! But here's the thing: on the scale of the power flows radiating in and out of this planet every second of every day, you the massive hurricane are even punier than the human facing the tornado; a mere swirly detail of barista foam art on the planetary coffee cup.

So if you have feels on this topic, I strongly urge you to do enough fact-checking and back-of-the-envelope arithmetic to test how realistic they are.
posted by flabdablet at 1:52 AM on January 9 [1 favorite]


We've already started to cook ourselves from making a tiny change to the Earth's ability to do exactly that.

From the CO2 emitted from dirty power, and very fractionally from the heat emitted from combustion. I think nickmark was making the point that having loads of clean power would not cause the same problem and we shouldn't shy away from growing it.
posted by biffa at 6:27 AM on January 9 [1 favorite]


would not cause the same problem

Quite so. As I said upthread, in the short term it would definitely alleviate them. But any energy-use strategy predicated on endless growth is going to cause horrible problems eventually; that's just inherent in the nature of endless growth.

So we'd be smart to take the present cutover from fuel burning to energy flow tapping, the one we're already in the early stages of and need to accelerate as a matter of urgency, and use it as an opportunity to re-think energy policy everywhere in such a way as not to predicate it on growth but on the twin goals of sufficiency and efficiency instead.

Even apart from the notable benefit of being actually sane, embracing these as policy goals would let the present cutover complete faster and cheaper.
posted by flabdablet at 7:06 AM on January 9 [1 favorite]


But any energy-use strategy predicated on endless growth is going to cause horrible problems eventually; that's just inherent in the nature of endless growth.

Again, I fully embrace this.

My job is in energy efficiency policy for a US utility; after years of focusing on regulatory reporting and "how are we going to meet our savings goals this year" short-term stuff I'm now in a role where I get to think about bigger-picture, medium-range stuff. One of the things I'm wrestling with is exactly the long-term role of energy efficiency as a utility resource, and the question that keeps coming up (on the electric side of the business, gas has other things to worry about) is, "once we've got all or even only most of the carbon out of the electricity supply [we're on track to be 80% carbon free by 2030], energy efficiency will have very little carbon savings to justify it, and electricity will have small and declining marginal cost of generation - so what is the public policy benefit to utility EE programs at that point?" (Yes, capacity will still cost, and so load management and peak-reduction will clearly have value - but once the solar farm is built, there's no additional cost from using it, like how you have to buy fuel for thermal generation.)

(There's an obvious related question of "should we still be billing by the kWh at that point", as flabdablet alludes to with the "too cheap to meter" comment above - but let's set that aside for the moment.)

notoriety public's comment about cooking ourselves with cheap clean energy was a little jarring and made me wonder if I was missing some important thermodynamic angle. But I'm not sure that's the case based on the discussion. If what notoriety public is talking about is basically the gradual buildup of heat from less-than-magically-efficient conversion processes, I don't think that's enough (on its own) to support a policy argument for efficiency, partly (but not only) because the "boil the oceans" scenario is a long-term doomsday built on an assumption of endless growth which cannot actually come to pass. Not that predicating our energy strategy on endless growth wouldn't have horrible consequences, but rather a) horrible consequences would start to show up long before the oceans boiled, and b) endless growth is a fiction anyway; it's just a question of whether growth ends in a managed leveling-off or a catastrophic crash.

I should say that when I refer to a "policy argument for efficiency" I'm referring to utility-funded energy efficiency programs that seek to use EE as a utility resource, and trying to find a rationale that says "this is why this is a cost-effective use of utility customer dollars." That's relatively easy when energy is dirty and carries measurable marginal cost, but it's more complicated in a too-cheap-to-meter scenario. If electricity truly becomes too clean and cheap to meter, it seems like it's probably too cheap to bother saving, and I'm trying to understand if there would still be environmental benefit from end-use efficiency at that point.
posted by nickmark at 8:11 AM on January 9 [6 favorites]


Solar PV only becomes too cheap to meter for a relatively small part of the day. To meet the general expectation of being able to operate any device at any time requires that energy generation gets time-shifted and that needs storage.

Storage costs money, and because the components it's most often made of are mass-produced cells whose individual storage capacities are very small relative to the amount of energy that needs to be time-shifted, those costs scale pretty much linearly with that amount of energy.

So if end-use efficiency is generally better in region A than in region B, region A's inhabitants will be paying correspondingly less for storage.

Similarly, a highly efficient car should cost less than an otherwise equivalent but less efficient model, simply because batteries are the most expensive part of an EV and the efficient car needs less of them to achieve comparable or even superior range.

Aptera, for example, is offering a 100kWh variant rated for 1,000 miles on a full charge; and regardless of battery, the design is efficient enough that PV cells incorporated in the bodywork will add more than enough charge to cover the majority of daily commutes just by parking the thing in the sun. Unless a car is that efficient, putting solar PV on it at all is not much use.
posted by flabdablet at 12:08 PM on January 9 [1 favorite]


Solar PV only becomes too cheap to meter for a relatively small part of the day.

Well that just means it's not too cheap to meter. I suppose you could offer "free afternoons" or somthing, but you still need a meter to know how much electricity not to charge for. And if you do, you're encouraging customers to shift their usage to time periods that are already the highest-load times of day, meaning you've got to build more generation to meet that higher peak demand.

Utilities are currently trying to get customers to charge their cars overnight (preferably in some staggered fashion so the chargers don't all start at the same time), rather than right away when they get home from work. Which touches also on the fact that electric utility peak hours tend to correspond to some degree to evening commuting hours, reducing the ability to use car batteries as a peaking resource (because the car is on the road). (I generally agree with your comments suggesting we should be putting more PV above parking lots, but I'm considerably less bullish than you - and many others, I acknowledge - on the idea of using cars for peak management or powering your home off the car.)

And don't even get me started on the terrifying topic of peak heating demands...
posted by nickmark at 2:25 PM on January 9 [1 favorite]


And if you do, you're encouraging customers to shift their usage to time periods that are already the highest-load times of day, meaning you've got to build more generation to meet that higher peak demand.


The UK has a goal of building 50GW of offshore wind by 2030, which reflects the fact that the UK has lots of good sites, is a global leader and can deliver electricity from offshore wind at wholesale prices on a par with the historical average price (ie much lower than current wholesale price). The UK has another 10+GW of onshore wind and looks to be about to start building more after a decade long hiatus. It also has a few GW of solar, but this is also growing. Current winter daily demand spikes at about 45GW (see gridwatch for example), so its entirely possible the UK will have enough output to meet all demand from wind alone at windy times, even in peak demand times. We can certainly expect to have way more wind at key times on offpeak times. What we might see is a shift away from the current model of supply following demand and towards a model where we encourage demand to move to adjust to available supply. That is basically what getting people to charge their EVs at particular times is about. What that will mean is that the current model with the largely predictable shape will change (ie the consistently recurring pattern on the gridwatch link will begin to change and may eventually be entirely different), since demand will be variable as well as supply. EV charging will not be the only shiftable demand, but it is currently likely to be the biggest.
posted by biffa at 4:29 PM on January 9 [2 favorites]


I suppose you could offer "free afternoons" or somthing

The electricity retailer I use does pretty much exactly that. Amber Electric makes its money not on electricity sales, but on a fixed monthly subscription; electricity I buy from them gets supplied to me at the same variable wholesale spot rate that my local distribution network charges all its retailers, with no markup for Amber. So they have no financial incentive to keep my usage high, and in fact are ideologically motivated to do the exact opposite.

They also offer ways for battery owners and EV owners to take the best advantage of variable wholesale prices.

Sometimes the wholesale spot rate goes negative and I actually get paid to draw energy from the grid at those times. I'm guessing this reflects the fact that winding a coal plant's output up and down in response to demand is hard enough that it occasionally becomes cheaper for coal plant operators to pay the grid to absorb their excessive output than to shut themselves down for a few hours.
posted by flabdablet at 3:05 AM on January 10


In the UK Octopus have been playing around with offering to pay consumers who make savings at particular times compared t their prior use: Savings sessions.
posted by biffa at 4:47 AM on January 10


« Older Catchy tunes on corruption   |   The Beaches -- Newer »


You are not currently logged in. Log in or create a new account to post comments.