Be careful with physics-backed statements! It seems many in this thread don't appreciate the energy density requirements of modern baseload. Don't want to crush their solarpunk dreams.
> It seems many in this thread don't appreciate the energy density requirements of modern baseload.
It's more about nuclear being too expensive. If nuclear's great value and doesn't need the state to underwrite everything from the disaster liability insurance on downwards, great, go knock yourself out; get finance, get approvals, build a plant, go sell your product on the market.
Just don't expect consumers to sign up for a multi-decade deal to guarantee to buy your output at a higher price than any other provider before you'll even pour any concrete.[0][1][2]
The cost of generation is not the same thing as the net cost of transitioning an energy grid to a different energy source. Wind and solar have cheap generation, but have much greater infrastructure costs than nuclear.
Storage is a huge one, we don't have effective means of storage besides hydroelectric dams which are geographically limited. Lithium ion batteries still take over $500/KWh to install (the <$200 figures are for the battery cells themselves, omitting the cost of installation, transformers, maintenance, etc.). And they're set to increase as raw materials are strained [1][2]. This is why plans to transition to a majority renewable grid typically assume that hydrogen, or some other form of storage will provide storage at a fraction of the cost of existing storage methods. Nobody has actually built commercial hydrogen storage, though, so this is a big assumption.
There's also the cost of transmission. Energy dense sources like nuclear power can be placed close to electricity demand. But low-density sources by definition need to be spread out and distributed. Decentralized generation is not a good thing, as it requires more transmission infrastructure to support. It's not uncommon for renewable projects to be denied because the infrastructure can't handle the transmission requirements [3].
Nuclear avoids these issues. It's a non-intermittent source with the greatest capacity factor of any generation system. Downtime is usually scheduled. This eliminates the need for storage. It's also energy dense. It can be used in place of existing fossil fuel heat engines, avoided in the need to make large build-outs of transmission infrastructure.
> but have much greater infrastructure costs than nuclear.
And proponents of solar forget that they have to cover large swaths of land with solar to begin with. Some places don't have the luxury of converting land to solar farms
Large swaths of land, but not absurdly large; the apartment towers two streets from me as I write, could generate about half their needs from PV cladding, and I'm in Berlin, slightly further north than the US-Canada border.
Likewise, PV cladding on electric cars, even though this is absolutely not the optimal place for PV, can generate 50-80% of the mean demand for that vehicle in most places.
But PV scales up and down, so you can also use it as a roofing material in a car park (one of the reasons cladding cars themselves in PV isn't the best idea), or on otherwise worthless land like the gap between the eastbound and westbound I-80 [0], or on top of reservoirs and rivers where we want to reduce evaporation [1].
Given where the sunlight mostly falls and where people mostly live, the biggest discrepancies are the UK (mostly as bad as the coastal mountains on the strip of Canada between Washington state and Alaska), and China (the people mostly live in the east which is relatively cloudy, the desert where they put some (but not all) of their PV is in the west).
For the small settlements that can't use PV and which it would be uneconomical to connect to a grid, there's wind and geothermal, and the rest may be so few as to not actually matter for environmental purposes if they do keep using diesel or methane. But probably not nuclear, because you can get a lot of transmission cable for the price of a reactor.
[0] I'm not sure how big that gap is, but a few random samples suggest 15 meters width, which means 20% efficient PV there has a nameplate capacity of 14 GW, and even assuming just 10% duty factor that's still more than a nuclear power plant, but the capacity factor is mostly going to be 18% in the area the I-80 goes through.
The entire US interstate network, assuming 15 m direction separation and 20% efficiency, has room for 235 GW nameplate, adjust accordingly for capacity factor.
There's much more length of railway line in the US, though I have no idea what the average width is over its length, and I don't understand the engineering constraints well enough to know how close PV could be placed to the rails (between the sleepers?)
[1] Lake Mead by itself would be enough for 128 GW nameplate, looking at the solar potential in that area it probably has a capacity factor of 22%.
The transmission and land costs also trade off against each other to an extent: land close to demand is more expensive, cheaper land far away from demand requires more transmission infrastructure.
> Nuclear avoids these issues. It's a non-intermittent source with the greatest capacity factor of any generation system.
A nuclear plant is pumping out electricity 24x7, but the problem is, no electricity consumer needs that kind of supply, certanly not at the prices nuclear needs to charge. When the wind is blowing and/or the sun is shining, every single electricity consumer would prefer to pick a cheaper, renewable, option.
The only way nuclear can even pretend to make the math work out is by going on and on about "baseload" and hoping that they can lock in buyers over several decades rather than having to actually look the spot price in the face.
Minimum energy demand is usually 80% of peak energy demand. Furthermore, the peak energy demand happens in the evening when solar tapers off. The time of day when energy is most scare is exactly when solar stops producing. Wind may or may not produce at the right time, as wind speeds through the day vary depending on location.
There's no "pretending" about base load. The vast majority of electricity demand is base load, and the peaks beyond base load happen when solar stops working.
It's more informative for judging value to examine peak vs min price. 80% production off peak just says there's some demand willing to exceed marginal production cost, which can be quite low.
Cost is a separate issue/argument from simply being able to provide baseload energy to modern society. Let's imagine we generate enough renewable energy to power modern society. What is the cost of overhauling our grid to dynamically distribute it in a way that avoids rolling blackouts? We're talking about reinventing our society to revolve around several transient energy sources. Nuclear stays on essentially 24/7, has enormous output, and the input cost is so insignificant that even if the cost of uranium increased 5x, it would barely change the cost of energy for consumers.
> What is the cost of overhauling our grid to dynamically distribute it in a way that avoids rolling blackouts?
The upper bound is the cost of a global power grid or a lot of storage, either of which would cost roughly half as much as the current annual spend on just unrefined crude oil.
> We're talking about reinventing our society to revolve around several transient energy sources.
No, that's just your failure of imagination.
> the input cost is so insignificant that even if the cost of uranium increased 5x, it would barely change the cost of energy for consumers.
The the reason for that is also why nuclear is the most expensive major power source right now: all the other things in nuclear power are much more expensive than the fuel.
> nuclear's great value and doesn't need the state to underwrite everything from the disaster liability insurance on downwards, great, go knock yourself out; get finance, get approvals,
Every significant solar energy installation in the world was built with massive government subsidies.
Energy density can be both important for transportation and also be important for the ability to accelerate energy production without a proportional growth in mining/resource harvesting that harms the planet.
A melon sized chunk of uranium powers an Aircraft carrier of the US navy for over 2 decades while it circumnavigates the globe hundreds of times, launching aircraft off its deck and powering remarkable levels of energy demand from on board systems.
For comparison, a single trip around the pacific for a conventionally fueled aircraft carrier costs 125 MILLION GALLONS of fuel.
It’s incredibly hard to wrap one’s head around what 20 years x 125 Million Gallons x Num_trips per year looks like.
I think you added some zeroes there. HMS Queen Elizabeth carries about 1.75 million gallons of fuel [1]. she has to fuel up about twice to sail 26,000 Nmi [2]. Pacific is about 8000 Nmi, so round tripping would be roughly a full tank.
> important for the ability to accelerate energy production without a proportional growth in mining/resource harvesting that harms the planet.
I agree.
> A melon sized chunk of uranium powers an Aircraft carrier of the US navy for over 2 decades … a single trip around the pacific for a conventionally fueled aircraft carrier costs 125 MILLION GALLONS of fuel.
Are you sure you didn't add a few zeros in there? I think the real energy density difference there is about a million, but you're at least 200 times more than that, depending on Num_trips?
(But yes, to the core point, transport is the one thing where energy density matters, and a nuclear powered aircraft carrier, or sub, is totally a thing where atomic power shines. Subs especially. Just that they're not a major part of the problem, and while this is a fun diversion I had been more interested in baseload here).
> Energy density can be both important for transportation and also be important for the ability to accelerate energy production without a proportional growth in mining/resource harvesting that harms the planet.
>A melon sized chunk of uranium powers an Aircraft carrier of the US navy for over 2 decades while it circumnavigates the globe hundreds of times, launching aircraft off its deck and powering remarkable levels of energy demand from on board systems.
You've made a bit of a mistake here, thinking that energy density and resource harvesting are always opposed. They're not. The aircraft carrier uses 93% HEU, while most reactors use uranium that is 10-20x less enriched. The same amount of earth has to be mined either way, but the HEU used in the core of an aircraft carrier or nuclear sub has to undergo a lot more expensive and resource intensive (and downright dangerous) post-processing.
> A melon sized chunk of uranium powers an Aircraft carrier of the US navy for over 2 decades while it circumnavigates the globe hundreds of times, launching aircraft off its deck and powering remarkable levels of energy demand from on board systems.
And how much does a civilian nuclear reactor weigh?
Even compare a solar panel (there are panels on the market that produce 50GJ/kg over their life, and it's almost all sand) to the fuel assembly plus a type A storage cask.
Even if we accept your ridiculous premise, solar wins.
Depends on the details. For land use, as I said in my other comments, the energy density is a red herring — the mass (and the area) aren't what we're limited by anyway.
For transport, the limits depend on the nature of the transport. Cars are fine with LiIon; aircraft can be fine with LiIon for a few hundred miles but not thousands; hydrogen from electrolysis can power aircraft or rockets but are a bad choice for submarines; PV in a spacecraft means an ion drive and not a launch, the higher specific impulse doesn't make up for the lower absolute thrust in that scenario; nuclear spacecraft would be great except everyone's terrified of it.
But none of this matters either way with the power grid.
> For land use, as I said in my other comments, the energy density is a red herring — the mass (and the area) aren't what we're limited by anyway.
My point is, even if we accept the broken premise then the conclusion is still to plan and fund as much solar and wind as possible until we've provisioned about 80% of net energy. At that point you consider the tradeoff between LCOS of whatever battery tech can be scaled and the cost of nuclear (or if we're still operating under the faulty premise, the extra land and mass the reactor will consume).
Just the fuel and storage cask of nuclear fuel weighs almost as much a solar panel for the same energy content -- the discarded depleted uranium weighs more.
If you look at low concentration Uranium mines like Inkai (which are already close to a majority and will be comparatively high yield if nuclear is expanded) you have hundreds of square kilometers of poisoned and unusable land and poisoned ground water that will probably never be properly remediated producing about 15W/m^2 . Even Husab which is open pit only produces about 180W/m^2 from the currently occupied land and 30W/m^2 from the whole strike.
Aren't you the one that brought up transportation? This is the most pedantic, exhausting thread I've ever seen on HN. So many surface level opinions that I feel like I'm on reddit
> The baseload[1] (also base load) is the minimum level of demand on an electrical grid over a span of time, for example, one week. This demand can be met by unvarying power plants,[2] dispatchable generation,[3] or by a collection of smaller intermittent energy sources,[4] depending on which approach has the best mix of low cost, availability and high reliability in any particular market.
Exactly, that is demand side. For the generation side coal and nuclear got the label "base load" plants, but that is simply a function of them being inflexible and that they used to be cheaper.
Nothing intrinsic to functioning of the grid, simply an economic consequence.
Solar and wind aren't 'flexible' though, they are unreliable. You don't get to choose when they start and stop, even for solar predicting sun during the day isn't reliable due to random clouds cutting out 80% of incoming light.
'Flexible' generation is gas and oil, as well as most storage systems. Either we need an enormous overprovisioning of both renewables and storage so that we can handle the 0.1% of cases (~= 1 day every 3 years), or we have something we can actually schedule when we choose, not when nature's chaotic systems choose for us.
PV is cheap enough that over-provisioning by a factor of a few-fold is basically a no brainer. Not sure where the crossover point is for long term (not just nighttime) storage or global grid, but a factor of x2 is cheap enough to just do first and ask questions later.
Overprovisioning doesn't help unless you distribute it wide enough that they don't all have the same points of inactivity. Even then wide regional issues can still give temporally correlated downtime, eg. Texas's wind turbines all not working at the same time due to low temperatures and icing. It doesn't matter how much you overprovision, 10 x 0 = 0.
Overprovisioning plus a wide regional distribution could work, but then you need lots of extra power transmission capacity, which is also expensive, and will be 80% idle 99% of the time.
Remember, we're engineering for the worst case scenario where we still need to provide power here. If we can think of it, it will happen sooner or later. Nuclear power plants are designed to be safe even if an airplane is flown into them, it's not good enough for renewables to then turn around and say "you need to redesign your entire society around our power being intermittent".
I'm hugely pro-renewables, but only for remote areas. For cities, wind/solar don't make sense due to reliability and energy density.
> It doesn't matter how much you overprovision, 10 x 0 = 0.
Yeah, but x0 only happens at night. I'm not sure what the multiplier is in a really bad storm, nor how long that lasts (~= night would be fine, you're already putting in that kind of storage).
> Overprovisioning plus a wide regional distribution could work, but then you need lots of extra power transmission capacity, which is also expensive, and will be 80% idle 99% of the time.
IIRC a global grid is (naïvely) ~= the cost of 6 months crude oil. Expensive in absolute money terms, but not relative money terms; though political cost is something else entirely.
x0 can also happen due to equipment failure, eg. if there was extreme heat that put all of the inverters into safety shutdown. Again, we aren't dealing with day-to-day, we're dealing with the 0.1% chance scenarios. These are what make renewables expensive when we're trying to supply high availability power.
The problem with the global grid is that it would need to be similarly overprovisioned so that during low-probability failure scenarios the few remaining power nodes could supply the entire thing. 10x overprovisioning here looks a lot more expensive.
Is trivially true, and applies to everything. It's not really worth bothering to mention because everything else also sufferers this.
> The problem with the global grid is that it would need to be similarly overprovisioned so that during low-probability failure scenarios the few remaining power nodes could supply the entire thing.
The need to over-provision a grid is obvious, but…
> 10x overprovisioning here looks a lot more expensive.
First: Why do that by a factor of x10? Best redundancy here is geographical diversity rather than a fatter… I was going to say "cable", but it is (or collectively, they are) the order of a few square meters cross section and that feels wrong as a name. But that thing is best spread out, not kept singular and made wider, whatever you call it.
Second: Even x10, the main limit is "that's a lot of stuff to mine, how do we reorganise the miners from coal and oil to metals" rather than the $ cost — while "a trillion" of anything is a lot for one person to contemplate, compared to the cost of what is currently dug up and then set on fire to provide the same power, it's quite cheap.
> Solar and wind aren't 'flexible' though, they are unreliable. You don't get to choose when they start and stop, even for solar predicting sun during the day isn't reliable due to random clouds cutting out 80% of incoming light.
They're both.
When they are able to generate power, they can turn on and off on a moment's notice. Nuclear and coal plants often need hours for a significant change in output.
It isn’t simply economic. I don’t think you understand the fundamental purpose of an energy grid.
It isn’t just a bunch of wires existing on their own. It is the energy delivery infrastructure for all of society.
“Baseload” is an attribute of the grid itself that indicates the minimum energy capacity these wires at present carry.
Baseload is NOT a constant value or a constant consumption pattern across time of day/day of week alone.
Baseload reflects the consumption of energy by society that the grid is DESIGNED to serve at any time.
In other words, were one to use your definition, the grid would no longer be considered a functioning grid anymore but one that is broken since it is incapable of meeting its minimum design specifications.
I think you misunderstand "base load". What you are talking about is likely peak load, like the problem of everyone putting on their tea kettles during football pauses. [1] In the UK they went with pumped hydro ~50 years ago to solve this.
It is a constant value. The rest used to be filled by peaker-plants or hydro. While slowly regulating the inflexible "base load" plants to follow the seasonal cycles.
There is nothing inherent to this definition that it must be slow inflexible plants that provide it. More interesting discussions comes from how do you provide system strength, frequency regulation and so on when you decrease the synchronous components in the grid, because those are actual hard questions.
For example, there is ongoing research in grid-forming inverters. This is what you do if you run your solar-powered home in island mode, and as anyone who has done it knows starting electrical engines sucks. It becomes a much more complex problem with destabilizing factors in continent-scale grids.
You are conflating too many popular terms incorrectly while trying to make your point.
Baseload as defined by Wikipedia that you cited does not contradict my point, you don’t seem to understand the nuance of what economic demand is and why that is different from an attribute or minimum design spec.
It isn’t the consequence as you originally declared, it is an attribute of the grid itself. That’s a very important thing to understand. Anything can provide that input to the grid, but that means the source providing input to the grid must meet that very basic design specification.
Flexibility of generation capacity coming online is easily compensated by other parts of the grid such as the storage or load shifting characteristics of the grid.
If you go to https://model.energy/ and look at the cost of a wind/solar/battery/hydrogen system for producing a cost optimized "synthetic baseload" source (using historical climate data), you will find that a 10x reduction in battery price is not needed to beat nuclear.
I’m not sure how these kind of over simplified look at technological progress are useful at all (except as a fun exercise and maybe as a tech tree in a video game).
In reality this is never this simplistic, and you actually run the risk of whitewashing history or whole industries. There are still very good use cases for wood energy (and there will be for all foreseeable future) while coal energy is pretty much just legacy at this point and will probably only be used recreationally in a decade or two. Natural gas on the other hand might get a boost with on-site carbon capture technology and might actually end up cleaner [in some cases] then nuclear or renewables + batteries. You also completely skipped hydro-power which has existed longer then coal and is still on a good run.
I doubt chemical batteries will ever be cheap enough for grid scale power. There are other cheaper forms of energy storage that work better in bulk. Batteries are mainly useful for being self-contained and dense. Neither of these are hard requirements for fixed grid-scale installations. Pumped storage is pretty cheap already in the geographies where it works, and geothermal storage has a lot of interest and potential at the moment.
Never is a long time. There hasn’t really been any demand for a chemical battery capable of large scale storage with frequent drain-recharge cycles. That is until we build out large scale renewable power plants. So if somebody has ever invented a cheap chemical battery that fulfills grid needs, that invention was ahead of its time and has been lost in obscurity.
So even if pumped hydro remains our best technology for large scale storage at the moment, I still remain optimistic that in a decade we will have market ready chemical batteries that rivals pumped hydro in places where geography does not favor the latter. I’m particularly looking at molten salt (or liquid metal) batteries here, with some storage facilities being under construction already.
Yeah, it can store 120 megawatt hours of power, which is pretty big, but the largest pumped storage battery in the US stores 24,000 megawatt hours of power and started operation 36 years ago.
Renewables don't need a 10x drop in battery price. They're viable right now to make up the large majority of the grid, with no storage, and are far cheaper and faster than nuclear. Variability is not an issue and it never was.
You should also look at the cost curve of batteries. It's linear decreasing on a log scale. A lot will change while we wait a decade for nuclear to be built.
This is just... incorrect. Bare minimum, you need massive upgrades to the grid in order to be able to move power around at night from places where the wind is blowing to places where it isn't (I've been told the wind is always blowing somewhere and it's always enough, but I'll believe that when I see it done at scale with no other power source). Keep in mind we'll likely be seeing a significant increase in night time power use to charge EVs as well.
In the absence of such grid capability (i.e. our situation today), on a windless night your renewable production is zero-- so you either need massive batteries, fossil fuels or nuclear.
Also, the fact that batteries have been getting cheaper isn't a guarantee this curve will continue, especially with lithium battery demand going through the roof and supply chain bottlenecks being likely. There are promising signs of lower costs with some of the larger, heavier battery types, but that's still in very early R&D stages and who knows how it will pan out.
If we're not adding more fission generation, I don't see how we can avoid burning a whole lot more fossil fuels for a longer time unless we make our peace with regular blackouts. Every other solution seems to involve banking on tech we don't have yet and don't know when or if we will.
We may need some gas peaker plants to occasionally run in the short-term, but they wouldn't be a significant fraction of energy generation. And the implication of that isn't that renewables leads to more fossil fuel usage than nuclear. You've also go to factor in that we can build out renewables a lot quicker.
This is the most straightforward way to progress energy generation, increase fuel energy density
Renewable energy will find its place after batteries decrease 10x in price