Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

It's a fair point that nuclear (and all power plants) need maintenance windows where they come offline (and occasionally unplanned outages). But this is not the same as saying nuclear is not dispatchable, that's just incorrect.


nuclear isn't dispatch-able for a different reason: you don't turn it off. Nuclear is relatively expensive, and those expenses are roughly 100% capex cost, so if you consider a reactor with a 10 billion construction cost and a 50 year lifespan, every hour you turn off the reactor costs at least ~$25k (or more if you assume the reactor was intending to do better than break-even.


If by dispatch-able you mean something that can cheaply fill the gaps when wind/solar is not producing, then nuclear is not dispatch-able. For that purpose, we really only have fossil plants (and hydro in a few areas)

But if you compare the predictability of nuclear to that of wind/solar, nuclear is a lot easier to plan, and also requires way less (if any) contribution from other sources.

Also, I would argue that the current prices for building nuclear plants is at least 2-4x higher than they should be (depending on location).

If the regulations for nuclear were to be scaled back to a point where the net average harm caused per GWh was just slightly less than for the alternatives, and if we allowed a free, competitive market for the construction, nuclear would become a lot cheaper than today.


> and also requires way less (if any) contribution from other sources.

This part is wrong. Electricity demand is varies about 50% over the course of the day and about 50% over the course of a year, so a 100% nuclear grid would only be operating at ~50% capacity which would double the costs.

On a local scale, renewables aren't predictable, but over large areas (e.g. US/EU), almost all of the variability cancels out (especially since wind and solar are anti-correlated with each other). Both nuclear and solar/wind grids (or a combination of both) will require some amount of over-capacity/hydro storage/battery storage/gas-peaker plants to economically provide consistent power. My guess is that we'll settle on some combination of over-provisioned solar/wind for seasonal variation combined with hydro/gas for daily and battery for hourly and faster variation. (and possibly using demand side shaping like desalination/water heating to use excess energy).

I find it somewhat hard to believe that nuclear plants could easily be made dramatically cheaper and easier to build if regulation wasn't a problem because if that were the case, we would expect to see China and India building lots of nuclear reactors cheaply. There's obviously further room for reactor design optimization, but I don't think it's as simple as just blaming regulation.


>> and also requires way less (if any) contribution from other sources.

> This part is wrong.

France used to provide 75% of their electricity demand from Nuclear. Add their hydro power, and it was 85-90%. I'm not aware of any other country reaching similar figures using wind/solar, ever.

Some countries (like Denmark) have surpassed 50% from wind/solar, but at least in the case of Denmark, that relies heavily on supplementing it with hydropower from Norway/Sweden.

> On a local scale, renewables aren't predictable, but over large areas ....

The continent wide grid capacity needed for this is not only expensive, it's also fragile. If you add the extra grid costs to wind and solar, it's no longer very cheap.

Nuclear also benefits from a grid, but rely a lot less on it than wind/solar does. Even if you cut off the grid (due to an EU breakup, let's say), countries with nuclear power would be fine.

> I find it somewhat hard to believe that nuclear plants could easily be made dramatically cheaper

Almost anything of that sort CAN be made a lot cheaper, as long as a free market is allowed to operate and economies of scale are achieved. On top of that, technological progress makes it possible to get more from less over time (including safety).

As for China and India, well China IS building a lot more Nuclear than anyone else (probably everyone else combined). Cost estimations for them are uncertain, but seem cheaper than Korea. (Korean prices are assumed to be ~$40/MWh).

I believe that there is a lot left on the table in terms of efficiencies to be gained if competition and innovation across countries were encouraged, but even if you chose NOT to believe that, well at least nuclear has shown that it can deliver up to 75%.

Wind + solar has not shown anything similar, at least not yet.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: