ITER is an experimental fusion reactor in development. According to Wikipedia, it's supposed to produce 10 times its input energy.

But according to New Energy Times, it seems to now be formally acknowledged that it will perform nowhere near that:

In an article in the French newspaper Le Canard Enchainé last week, Michel Claessens, the former ITER organization spokesman, explained the ITER power discrepancy. “For many years, it was claimed that the reactor will generate ten times the power injected. It is completely wrong. Thanks to a patient investigation, the American journalist Steven Krivit showed that ITER will consume as much [power] as it will generate,” Claessens said. “We know now that the net [power] balance will be close to zero.” The newspaper asked the ITER organization for a response. The organization sent an official but unsigned response, provided under the direction of the current ITER spokesman, Laban Coblentz. “It is obvious that all the systems of the ITER installation will consume more energy than produced by the plasma,” the ITER organization said.

You can see this cited in other papers, such at this:

There’s an obsessive guy named Steven B. Krivit who seems to spend most of his time looking at various alternative nuclear generation technologies, including debunking cold fusion. His piece from November 3rd, 2021 asserted that he’d identified in 2017 that ITER wouldn’t be generating more energy than was put in, and that ITER finally admitted it to a press outlet.

Is it true that "the ITER Organization's" position on the experiment is that ITER will not produce energy sufficient to cover input?

  • Can you rephrase this question? You only identify a "former spokesman," not the organization itself. Please make it clear what you are asking for confirmation of.
    – Avery
    Nov 18, 2021 at 19:49
  • @Avery that's a quote. "The organization sent an official but unsigned response" I'm wondering if this is their position now on the projection of ITER's energy output. Nov 18, 2021 at 20:01
  • 9
    I'm not clear on what the controversy is about. Iter is a research project with no means of capturing the thermal output and converting it to electricity. The goal is to demonstrate a thermal output of 500 MW from a thermal input of 50 MW - a 10:1 ratio, which will far exceed the 0.7:1 ratio which is the highest yet achieved. Has someone claimed otherwise?
    – Mark
    Nov 19, 2021 at 2:33
  • 3
    Sabine Hossenfelder has a good video about the difference between Q_plasma (which only takes into account the energy that goes into the plasma, and is the one that's ITER hopes to get to 10) and Q_total (which takes into account all energy used by the fusion reactor, and only looks at captured output power, and is therefore much lower -- still below 1). Nov 19, 2021 at 20:24
  • 1
    @Mark I think the question is quite clear and elaborated on thoroughly in the OP's second link with many examples. The common understanding is that ITER could, at least in theory, provide a net gain of electricity, not just heat. If you already understood that was not the case, then you are ahead of most people, including myself before reading this question.
    – Harabeck
    Nov 19, 2021 at 20:33

1 Answer 1


There are two different statements, which seem to be contradictory, but are simply saying different things:

  • ITER aims to produce 10 times as much heat energy as put in
  • ITER will consume more total energy than it generates

The confusion comes from a simple misunderstanding: heat energy is not the only input - for instance, power is needed to cool superconductors; nor is the heat energy a directly usable output - to produce electricity, it would be used to produce steam and turn a turbine. As a research device, ITER is neither designed to capture the energy produced, nor to minimise the total energy used to create the plasma.

As a very rough analogy, picture an internal combustion engine which requires energy to run fuel pumps, starter motor, etc; and which will eventually produce useful energy via a drive train and a dynamo. In a prototype, you can ignore the efficiency of those "external" components, and study how well the engine extracts energy from the fuel. Current fusion reactors are like an engine where the fuel ignites only while the starter motor is running, because it doesn't turn the engine fast enough to sustain the process; ITER's aim is the equivalent of using the fuel to turn the engine faster than the starter motor, but with no drive train attached.

The Wikipedia article (permalink from the time this question was posted) summarises the aim thus:

to achieve enough fusion to produce 10 times as much thermal output power as thermal power absorbed by the plasma for short time periods

And goes on to say:

ITER's thermonuclear fusion reactor will use over 300MW of electrical power to cause the plasma to absorb 50 MW of thermal power, creating 500 MW of heat from fusion for periods of 400 to 600 seconds. This would mean a ten-fold gain of plasma heating power or, as measured by heating input to thermal output, Q ≥ 10. The European STOA Fusion Project cautions that this figure refers only to the energy of the plasma itself, and that practical capture of this energy for electricity production would introduce significant inefficiencies which ITER is not designed to overcome.

The ITER website confirms this in an FAQ "What is the difference between plasma energy breakeven and engineering breakeven?":

Plasma energy breakeven is the moment when the efficiency of the fusion reaction reaches Q = 1 (...); that is, when the total fusion power produced during a plasma pulse equals the power injected into the systems that heat the plasma.
Engineering breakeven would take into consideration all of the plants systems—and not just external heating systems—in the evaluation of the input/output power balance of an electricity-producing fusion power plant. Commercial fusion plants will be designed based on a power balance that accounts for the entire facility [...]

The significance of the Q=10 target is not that it makes the design a useful source of power in itself, but that it allows study of "burning plasma":

When heating by the helium nuclei ("alpha heating") is dominant (over 50 percent) the plasma is said to be a "burning plasma."

The net energy gain is in the plasma, allowing the reaction to be somewhat self-sustaining. That gives the reactor properties that haven't been achieved before, and allows study into how future devices could operate.

  • Where does this 500 megawatts of heat go?
    – Abdullah
    Nov 20, 2021 at 7:25
  • @Abdullah As I understand it, it simply escapes into the room where the reactor is installed. Bear in mind that the reactor is only going to be run for a few minutes at a time - long enough to prove it works, and test certain processes inside it.
    – IMSoP
    Nov 20, 2021 at 10:27
  • 1
    @Abdullah: It's standard to pump cooling-water through a hot process to keep its temperature at a desired point. If that's what they do, and say they heat water from 20degC to 45degC (random numbers), then they'd need to pump about 5 tonnes of cooling-water per second to absorb the heat.
    – Nat
    Nov 21, 2021 at 13:12
  • 7
    @Abdullah: Their site says the combined cooling-water flow-rate'll be about 33 m^3/s.
    – Nat
    Nov 21, 2021 at 13:23
  • 3
    @Abdullah which, for the non-metric people out there, is 33 metric tons per second. That's a LOT of water to carry away a LOT of heat. In a commercial powerplant that heat would be used to generate steam to drive turbines, in this prototype it's going to be left to escape in a cooling pond or cooling tower, at least initially.
    – jwenting
    Nov 26, 2021 at 9:07

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .