Hacker News Reader

> So you need a really really stable base load, and one that you can turn up a little and down a little whenever you want: to account for the wildly varying input from wind and solar.

But nuclear isn't very good at quickly and continuously ramping up and down. It wants to run at mostly constant output[1]. This makes it a poor companion to renewables.

[1] Except for the couple of weeks a year when it's down for maintenance, refueling, or its river's water has gotten too hot or there isn't enough of it, then the output is zero. Actually, it's negative because you need to provide it with energy or it self-destroys.

Yes, you need a quicker power source to balance, and Sweden got that with the hydro power.

The hydro power is not enough as base power for the large cities in the south, on the other hand, in part because in part the capacity is not enough, and in part because how the extremely long transport lines from the far north (where the hydro power is located) to the major cities in the south are a serious bottle-neck. Increasing their capacity (across up to thousands of kilometers) isn't easily done in a few years even.

It's not an engineering problem either.

The thing is though, a lot of heavy industry is built in the North to take advantage of the cheap power there. Shoring up the grid connections would increase the electricity rates in the northern region, due to competition from the South.

Which is why you need gas turbines for minor fluctuations. Buuut if you don't have any nuclear it's gas all the way.

EDIT: and yes, if the hydro is far away, it doesn't help much. Besides hydro also depends on weather, although on periods about 10 years or so. But when just once you don't have enough water then what?

>But nuclear isn't very good at quickly and continuously ramping up and down. It wants to run at mostly constant output[1]

Modern reactors can actually ramp up for down pretty quickly, as far a the grid is considered. That is to say, they can go from 50% to 100% in an hour or two.

However, you want to run them mostly constantly because once you build them, they are basically free power.

Besides, Constant power IS what you want for a "stable base load". You can start and stop solar and wind basically at the push of a button for very rapid response.

The point of Nuclear or fossil fuels with respect to the grid is reliability.

> You can start and stop solar and wind basically at the push of a button for very rapid response.

No, you can't do that with solar and wind, that's the whole point. Sure, you can choose to waste the renewable energy when it's in excess, but you can't conjure it when it is not.

If you want a grid that has a significant renewable percentage, you don't want stable base load plants, you want power storage, peaker power plants and demand scaling in response to supply.

Cue someone scoffing and saying, yeah, exactly, that's why we don't want any significant amount of renewables on our grid -- Fine! At least that's honest.

lets be clear, for your a grid, you have a base load that is basically constant, and then variation on top of that. You can think of as a base load and a variable load of another similar magnitude.

The ideal source for base load is something that has an uptime as close to 100% as possible.

Variable load is obviously the tricky part, because that is where you want storage to time shift, overbuilding to deal with seasonal changes, and build redundant peaker plants. This is the really expensive part, far more expensive than building nuclear

> But nuclear isn't very good at quickly and continuously ramping up and down

In terms of kW/min change rate faster then most gas peaker plants used all over the world.

It is a totally different question if that is economical (the costs of a nuclear plant don't really change much when running at 20% or 100%). So most operators run their plants at maximum if possible just like what wind/solar do when it is windy/sunny. France is the big exception to this due to how large portion of their production is nuclear they have to follow the load with some of the reactors.

Sweden doesn't use coal to any appreciable degree.

Hydro provides about 42%, nuclear 29%, wind 21%, thermal power 5.8%, and "unspecified" at 1.9%.

(Thanks to HN user fifilura who pointed people to the dashboard at https://www.svk.se/om-kraftsystemet/kontrollrummet/ a few months ago, showing the real-time data and breakdown.)

Yea. You need to constantly produce the same amount of power that is being consumed, or you get a fall of the frequency in the system. Wind doesn't care about that. You don't need to look long at a graph showing the typical power production of wind vs something like nuclear over a day, to see that wind power is not only extremely unreliable. It also requires massive balancing measures from the other power types in the mix, which is sometimes easier said than done.

You definitely want to let the wind power mostly "top up" the mix, and not constitute major parts of it, unless you have tons of something else to balance up the fluctuations with (related, nuclear is not great for balancing this, as it can't typically change output quickly, but is great as the base load, just as the parent comment says).

Speaking of topping up which I agree with, I think it bears saying here that today with unexpanded nuclear in Sweden, wind accounts for only about 20% of energy production. Or if only including renewable sources NOT including nuclear, wind still only accounts for about 30%.

Nuclear power "only" accounts for 29% of total production. Wind and especially offshore wind has huge potential to expand, but the government has taken an obstructionist stance towards that mode of power production. It doesn't make sense cutting wind subsidies amounting to a few billion while guaranteeing 400 billion to nuclear investments.

There has already been challenges in balancing out the extreme fluctuations from wind with the current amount (leading to the need to start gas turbines which were meant for emergency power multiple times over the last few years, and even oil plants), and expanding it would only make this worse, as there would not be enough easily balancing power sources close by (hydro can be used to balance to some extent, but are hundreds to thousands of km away, with limited capacity in the extremely long power lines, meaning that it can't account for all of that).

I think society will have to adapt to wildly swinging rates. Earlier in history, per minute metering was just not possible, but today it is. It's time we make use of that.

Not quite true as the power meant for emergencies has not been activated. Oil power plants has been in use because oil prices have been low while electricity prices have been high. That the price of electricity fluctuates is a feature of the system not a drawback because it shapes consumer demand according to availability.

The balance is built into the system. When elecricity prices are low wind power just disconnects generation capacity, while nuclear has to sell electricity for negative prices.

Firstly, the fluctuations happen on a timescale (hours, not days) where price fluctuations don't really have a chance to influence behavior via the market (which is mostly dealt with the day before). In particular heavy industries which are the bulk consumers, can't typically change their production plans on the hour, while on the other hand a wind front might arrive one hour late or early, causing a huge deviation in the production plan, that must immediately be balanced.

And this leads to the second point: There needs to be enough of available balancing power to be able to handle even the worst such deviation to keep the balance, as otherwise a breakdown of the whole system can happen, which would take days to restore from (yea, you can do emergency operations like closing down industries or apartment areas to keep the system from breaking down, but that in itself can have huge economic consequences if nothing else).

Why wouldn't price fluctuations influence consumer behavior? Electricity is purchased in bulk in Nord Pool whose prices fluctuate by the hour. Consumers, who account for 58% of production, certainly regulate their consumption based on prices. Even industries do and there have been several high-profile examples of paper mills reducing their production due to high electricity costs. Wind availability and temperature is what determines electricity prices, neither of which is particularly hard to predict.

Regarding your second point, this is handled via imports and exports and standby capacity. The Nordic power grid is getting denser which reduces the chances of catastrophic failure. While the Swedish right-wing press has predicted brownouts and "electricity ransoning" ever since the right-wing party made more nuclear part of their election platform back in 2020 (what a coincidence...), those dire predictions has not come true.

> Wind and solar are terrific but unreliable. What if it’s not windy or sunny?

Battery storage can fill the gap. This blog post makes a good argument why battery storage and over provisioning of renewables could solve this problem - https://caseyhandmer.wordpress.com/2023/07/12/grid-storage-b.... In addition it doesn't include dynamically controlling demand (for example not charging EVs when demand is high).

> So we need something to replace coal, which provides the grid’s base load. That thing can basically only be nuclear at this point.

Sweden doesn't appear to use coal as baseload. The UK did, but managed to replace most of it with gas, wind, solar, reduced demand and increased imports (https://www.mygridgb.co.uk/historicaldata/).

> and increased imports

From France, who has a shit-load of nuclear. :-)

I'm for more nuclear (a decent amount is generated in the UK), but I think your statement:

"So we need something to replace coal, which provides the grid’s base load. That thing can basically only be nuclear at this point."

doesn't take into the impact storage (both stationary and in the form of EVs) will have in the future.

> the impact storage (both stationary and in the form of EVs) will have in the future.

Have you ever heard the expression "hope is not a plan"?

I'm basing this on the huge increase in battery production.

From https://www.iea.org/data-and-statistics/charts/annual-grid-s...:

* Annual grid-scale battery storage additions went from 0.76 GW in 2016 to 11.21 GW in 2022. I couldn't find the GWh figures, but this is a fast ramp up.

* "In September 2022, India released its draft National Electricity Plan, setting out ambitious targets for the development of battery energy storage, with an estimated capacity of between 51 to 84 GW installed by 2031-32."

* "In December 2022, the Australian Renewable Energy Agency...announced funding support for a total of 2 GW/4.2 GWh of grid-scale storage capacity, equipped with grid-forming inverters to provide essential system services that are currently supplied by thermal power plants."

From https://www.iea.org/reports/global-ev-outlook-2023/executive...:

* "Automotive lithium-ion (Li-ion) battery demand increased by about 65% to 550 GWh in 2022, from about 330 GWh in 2021"

* "August 2022 and March 2023, major EV and battery makers announced cumulative post-IRA investments of at least USD 52 billion in North American EV supply chains – of which 50% is for battery manufacturing, and about 20% each for battery components and EV manufacturing."

These pages also list the challenges faced, so I'm not assuming it's a smooth road ahead. However, the trend is clear. Hopefully, technological improvements will continue and drive down costs further (I have no knowledge if this will happen).

I hope there's improvements in nuclear technology that allow for a faster, cheaper nuclear rollout and belief it has it's place.

But to return to original comment I replied that states

"we need something to replace coal...that thing can...only be nuclear at this point"

dismisses the increases in investment and production of battery storage.

No, they don't. The French grid wouldn't have made it through 2022 without a lot of imports from Germany, as too many reactors were down for maintenande. They found a fault which concerned most reactors of that type. On top of that, in the summer the cooling of the reactors is a growing worry with raising temperatures. Of course, you could decide to sterelize the rivers used for cooling. Alternatively, you have to throttle the nuclear output.

The kind of storage we would need in the Nordics is around 1 week of total consumption. That is roughly how long the longest really cold windless periods last in the winter (basically a large high pressure zone parking itself over the area and they can be big enough to cover pretty much all of Sweden) and sun is kinda bad during the winter this far north.

In the north thermal energy storage is a viable solution, and those can store energy for months. You can dump heat into them in summer and still pull from it in winter.

Since heating is by far the biggest consumer of energy in winter, just taking care of that with stored thermal energy will deal with a big fraction of the energy storage needs.

We need more districting heating systems.

There's several projects ongoing to store heat from district heating systems, for example in Västerås [1] and Linköping [2]

[1] https://www.malarenergi.se/om-malarenergi/framtidens-samhall... [2] https://www.svt.se/nyheter/lokalt/ost/de-lagrar-varme-i-berg...

We are talking about 2 to 3 million MWh for Sweden alone (~120 million MWh yearly consumption so in reality probably a bit more then that as the consumption is higher in the winter).

The largest battery based energy storage in the world is 3000 MWh so we just need to build thousand of those (or a 1000x larger one). Good luck with that.

The largest heat storage under construction is 90000 MWh so that is a step in the right direction but still falls short by around 35x and even more if we have to turn that heat into electricity (While a lot of this consumption is district heating it is not all of it. And a large chunk of new district heating is just massive heat pumps)

And now that you have built these tools to make wind/solar work you have to factor the price of these into the price of wind/solar and suddenly they do not look that cheap when compared to nuclear. This is why the government and Vattenfall in Sweden want to build more nuclear.

Realistically the only cost effective energy storage on this scale we have available is hydro and pretty much all the places where it is cheap in Sweden have already been built.

In general I think Sweden is on the right track. They have a lot of wind and are building a lot more of it. They can also build more nuclear at the same time.

Though Sweden (just like here in Finland) are slowly reaching the point that when it is windy the wind farms don't really make much money as the price on the spot markets is very close to 0 so we also need a lot of new intermittent energy consumption (industrial processes that can be slowed down/ramped up cost effectively, make hydrogen, fill the heat storages with heat pumps, etc). This is the place where batteries can play a big role by being able to "move" this electricity into the future to a moment where there is more consumption/less production.

I don’t think solar is as bad as you think it is during winter. During a stationary high pressure, you’ll have clear skies. Solar cells are more efficient when cold, and in fact in warmer countries the heat of midday can make them so inefficient that solar panels generate the most power in the morning and late afternoon.

It is bad due to the short days and low angle of the sun.

For example in Stockholm you get around 6h of daylight. Malmö (can’t really get further south than that) a bit over 7h.

Basically the northern most point of mainland US is a lot further south than the most southern place in Sweden. Thinking about how good solar panels are in the winter in Anchorage gives you some idea (which has slightly slightly shorter days then Stockholm at around 5h30min at their shortest)

6 hours of daylight, then you need to over-provision the solar so you can charge 18 hours worth of batteries in that time. This is still cost competitive with nuclear

It is way worse than “6 hours of daylight” due to the lower angle of the sun.

I don’t know the numbers for southern Sweden but for Stockholm level you get maybe 5% to 10% of the production you get in the summer.

Here in Finland we have it even worse but I know where to find the numbers.

https://www.fingrid.fi/en/electricity-market-information/sol...

Basically if we are lucky we hit 5% of a typical summer day for an hour maybe two. As the days are less then half as long you end up with ~1% the production vs summer.

But in reverse the panels are really good during the summer as the days are very long. This is why people bother installing solar here.

Is it really? Including the batteries?

And we're still ignoring the fact that many days you'll get barely any direct sunlight during those 6 hours of "daylight"

Our (Swedens) base load is mostly nuclear and hydro currently, the capacity generated from each type can be viewed live at https://www.svk.se/kontrollrummet

But we have multiple large industries starting up in the coming decades which will require substantial power, while at the same time electric vehicles are slowly but surely becoming mainstream, further straining our grid.

Increasing nuclear is a great (and necessary) decision, it's just a shame the decision wasn't taken 20 years ago.

Electric vehicles is both a bane and a boon. Let them charge when it's cheap, program them to sell back a fraction of the charge when it's expensive.

Some of the planned industry investment is hydrogen generation, which is perfect for varying power. The gas caverns can be filled when rates are low.

Yep, this (EVs with bidirectional charging) is something we have to do, as it could be used to balance the grid when necessary, diminishing the need to spin up fossil fuel power in these circumstances.

Unfortunately most EVs don't have the hardware to be used as batteries through their Type 2 (Mennekes) connector, as it requires a built-in inverter that just isn't there.

There are some new products (e.g. [1]) coming out that uses the CCS2 port on the car, providing the necessary inverter externally, which would enable older EVs to be used as batteries for grid balancing or other similar uses.

1: https://wallbox.com/sv_se/quasar-2-laddare-dubbelriktade-elf...

That is not what base load is. Base load is about some power plants that needs days to weeks to start and stop. They are unable to follow load changes. You build them anyway because they cost much less $$$ to run, but you need non-base load to handle varying power demands as your base load doesn't throttle very well.

While wind and solar do not meet the definition of base load, they clearly are in the spirit of base load.

When doing wind and solar you need non-base load, not more base load.

https://duckduckgo.com/?t=ffab&q=base+load&iax=images&ia=ima...

This keeps being repeated but it's nonsense, there are many technical and economical ways to reduce the auxiliary baseload requirement to zero or very close to zero. These include: HVDC to couple geographically distant regions, storage, flexible consumption, reduction of baseload consumption, hydro power for baseload, pricing based on source etc.

Sweden has a large amount of hydropower as baseload though.

The hydro power is located in the far north(west), meaning that there is a bottle-neck in the hundreds to thousands of km of transport lines needed to bring it to the big cities in the south.

The capacity of these lines is simply not enough to supply the cities in the south with only hydro power base load, which is why Sweden had the four major nuclear power stations built in the south, in the vicinity of major urban areas.

It was a really well thought out system, that has been degrading over the last years for political reasons.

> What if it’s not windy or sunny?

Sweden is 528 000 km² and there is part a geographical spread even into sea which use to have weather patterns not following those at land, part complemented heavily by hydro, and part large battery parks already in use to complement wind.

What kind of issues still remain after all these countermeasures? Still requiring countermeasures on the extremely costly scale of a massive nuclear expansion?

See my post, which I may have updated since you wrote this.

Renewables, without massive grid storage to dampen their unreliability can not be the base load upon which your grid depends.

This storage could exist, but we haven’t built it and it isn’t proven.

It is absolutely built and proven in numerous countries and more are planned / under construction by Vattenfall alone.

https://group.vattenfall.com/what-we-do/roadmap-to-fossil-fr...

After doing all of that, your wind and solar are no longer cheap. The cheap numbers per kwh for wind/solar currently are under the assumption that you take the power when you get it.

It doesn't matter if it's more expensive. The question if it eventually gets more expensive than nuclear, an energy form Sweden/Vattenfall recently turned away from because it was not financially feasible anymore.

I've seen cited that the cheapest chemical battery storage is similar to the price of gas generation. If we're considering LCOE numbers, yes, that would be more expensive than nuclear then. Your local regulatory environment notwithstanding.

Sweden has often very long transport routes for the energy, meaning that it is still a big challenge to supply the big cities, especially in the very south, with stable power, as the long transport routes don't have capacity enough for the increasing demand from an increasingly electrified society (charging electric cars etc).

As other comments are on to, you can't rely on wind for base energy due to the extreme fluctuations in output. It isn't uncommon with a few days of basically no wind at all in the whole country, typically when it is at its coldest, with temperatures sometimes around -30 - -40 C in the north, and perhaps -10 - -20 C in the south. Then you gotta have stable energy, or people's lives might be at risk.