Hacker News Reader

Comment by orangeoxidation | Hacker News Reader

Yup, that seems a strange argument. Power usage varies during the day, but it doesn't really go below some "base load".

Nuclear can run basically 24/7, but you cannot turn it on or off quick enough to to react to hourly changes in demand. So nuclear power is only good for base load.

We need peaker plants to get the rest. Gas plants are the popular (cheap) choice for this. Carbon free alternatives are pumped hydro or batteries.

With renewables the production capacity is variable as well.

To fill possible "holes" in it we don't need more base load. What renewables need are ... peaker plants. A role nuclear reactors are exceptionally unsuited to fill.

View on HN

> Nuclear can run basically 24/7, but you cannot turn it on or off quick enough to to react to hourly changes in demand. So nuclear power is only good for base load.

That's not exactly true, you can build nuclear plants for load following which provides some amount of flexibility, at the cost of some efficiency (about 1% I think).

IIRC French plants can operate between 30 and 100% rated power, and ramp rates can reach 5% per minute (though normal rates are 1 to 3). French nukes regularly have to ramp up and down quickly to compensate for wind variation and monday pickup (electricity consumptions goes way down over the weekend, especially nice spring weekends, then back way up on week start).

The less they produce, the higher is the cost of each kWh. But they are already not competitive at 100%.

Ah yes, the “heads I win, tails you lose” gambit, thank you for your very useful contributions.

Do carefully avoid looking at french electricity prices and compare them to germany’s too.

I doesn't make the cost structure of nuke power. It is how it is without any outside help.

> but you cannot turn it on or off quick enough to to react to hourly changes in demand

To be precise, it cannot stop consuming fuel quick enough that there's any important savings, and so nuclear power plants always want to run at full capacity because otherwise they are wasting fuel.

But more than that, fuel consumption is simply not a very big part of the running cost of a nuclear power plant, most is fixed cost so that even if they could change the fuel consumption quickly there's just not enough savings to really bother with it.

> To be precise, it cannot stop consuming fuel quick enough that there's any important savings, and so nuclear power plants always want to run at full capacity because otherwise they are wasting fuel. But more than that, fuel consumption is simply not a very big part of the running cost of a nuclear power plant, most is fixed cost so that even if they could change the fuel consumption quickly there's just not enough savings to really bother with it

(Sorry) but that sounds like a slightly long-winded way of saying that nuclear just isn't economically viable.

"the costs of renewables continue to fall due to incremental manufacturing and installation improvements while nuclear, despite over half a century of industrial experience, continues to see costs rising"[0]

[0] https://www.pv-magazine.com/2020/09/24/nuclear-power-is-now-...

Correct. We know nukes are not economically viable. They get less so with every passing day. Starting any new ones under the present circumstances would be extremely stupid, as no one would pay for power from it at a price it could offer.

> you cannot turn it on or off quick enough to to react to hourly changes in demand

Surely you'd want actually want suppliers to react to changes in spot price, not just demand? If it's windy and sunny, it might not matter if demand is high! If it's calm and cloudy, you have a problem.

If the spot electricity price is high, you want providers to jump in and supply electricity. If it's low, or indeed goes negative[0], you want them to shut down.

Nuclear just doesn't fit this model, since investors appear to want the strike price guaranteed for several decades before they'll even start pouring concrete for their plant.

[0] https://www.sciencedirect.com/science/article/pii/S266679242...

Reply on news.ycombinator.com