Hacker News Reader

Comment by adrianmonk | Hacker News Reader

I'm not an expert, but I'm not convinced about it reducing the need to maintain a nationwide grid.

Nuclear reactors take ages to ramp up and down. It's basically going to be generating the same amount of power 24x7, but demand is going to fluctuate. The more other areas you're connected to, the more opportunity there is to send that power to someone who can use it.

Obviously there's a law of diminishing returns at some point, so maybe the grid doesn't need to be as large as possible.

There are alternatives like energy storage (batteries, etc.), but you'd have to compare all the costs and benefits.

View on HN

Nukes are slow from a cold start but they can throttle up and down with load.

https://en.wikipedia.org/wiki/Load_following_power_plant#Nuc...

Modern nuclear plants with light water reactors are designed to have maneuvering capabilities in the 30-100% range with 5%/minute slope.

Wow, that's pretty fast. So 12 minutes worst case. TIL.

Still, I wonder if the economics don't favor running near 100%. You've already paid the high up-front cost of the equipment, and fuel costs are low, so I assume you're better off selling excess power when possible.

Of course. Though with increasing deployment of intermittent sources like wind and solar, occasionally the wholesale price can drop lower than the marginal cost of the nuclear plant (or even negative, if said power sources receive production subsidies regardless of the wholesale price), in which case it makes sense to throttle down.

Future grids with more intermittent renewables will have increasingly volatile prices, but not necessarily lower on average. So generators that can produce during high price periods (e.g. if wind and solar aren't producing much) can make a lot of money then, compensating for less income during low price periods.

I'd be willing to bet there are some significant thermal efficiency losses at 100% that would push you down closer to (made up number incoming) 80% at idle

What's the maneuvering ability of a coal plant? I looked at the article, but didn't see it.

Gas turbines are the ones that ramp the fastest and are used to handle dynamic peaky loads. Nuclear/coal are generally used as the baseline power because they're more efficient.

Fastest is a coal/Gas/Oil fired Steam Turbine driven power station sitting at full pressure sitting as spinning reserve

> Nuclear reactors take ages to ramp up and down.

Naval reactors are apparently very fast in this regard, so it's evidently not an inherent property of nuclear power.

(The USN also has an unparalleled record of safely operating reactors; more that 5,000 reactor-years clocked without major incident.)

> Naval reactors are apparently very fast in this regard, so it's evidently not an inherent property of nuclear power.

Ramping up is easy, ramping down... not so much.

It's not as much of an issue if you swim in your coolant, hence naval reactors are unique in being able to ramp both up, and down quickly.

Most civil reactors are next to lakes or rivers so it seems like they should have plenty of water. Is it a matter of civil reactors using smaller[cheaper] plumbing and pumps? My intuition is that Naval reactors are built to be very responsive because that's useful for naval applications, but civil reactors aren't built like that because that ability has less relative value for civil reactors.

Does it count that a Navy man was killed at SL-1?

https://en.wikipedia.org/wiki/SL-1 https://www.findagrave.com/memorial/67718734/richard-carlton...

SL-1 was an Army reactor, so I would say it doesn't count. I'd also not count the USS Thresher or the USS Scorpion, since reactor accidents weren't the reason those sank.

Relative to how many people die every year from wood and coal smoke? I would say it doesn't count in a relative sense.

> Nuclear reactors take ages to ramp up and down.

Run them at full blast and dump the extra energy into direct air carbon capture? Of course that would require building the CC plants but it could be planned for.

That's the whole point of creating grid-scale storage - giant batteries, lots of smaller batteries, pumped storage, pressurized air storage, traction railways, etc - or smarten up loads like EV charging so they can ramp up/down to soften the fluctuations in demand from other sources.

You'd probably just connect carbon capture plants to the grid, and run them when electricity is cheap for any reason (presumably any fossil sources have no reason to run at this time...)

On a grid with enough opportunistic loads, you'd never need to ramp down reactors.

I can see that if these are distributed, the grid can be more of a mesh than spokes on a wheel.

solar can do the same sort of thing.

Reply on news.ycombinator.com