Hacker News Reader

Comment by yodelshady | Hacker News Reader

The ~25 GW nameplate capacity of wind power in the UK is producing 0.76 GW as a type this.

Last calendar, it produced > 1 GW for 784 hours.

In the last four months, assuming a 40% capacity factor and therefore an expected capacity of 10 GW, there have been at least three periods with > 100 GWh shortfall.

The UK does not yet, but plans to, shift heating loads onto the electrical grid, meaning that is likely to become a 100 GWh shortfall in heating.

Please, show me an estimate for an alternative fossil-fuel system that can guarantee that shortfall.

edit: more specific to SMRs, and I have ranted on this before: they've been given ~ £ 500 M, which does... just feel like the right amount. We're all aware how 100 times that can be disappear into consultants, and much less simply wouldn't get you a complete product. £500 M I can see engineers spending.

View on HN

You are not wrong. More storage is needed; more solar is needed; more connectivity to Europe is needed. The wind not blowing is a local effect. It's more rare offshore. It's very rare for it to be continent wide. So connectivity helps, storage helps, having people with solar on their roofs helps.

There's a need for "base load". I hesitate to use that term because people get carried away with weird notions of it. If you slap a percentage on it, it's probably closer to 5% than to 50% of overall capacity if you look at it at a European level. We're talking a cloudy, wind free day across the entirety of the continent. A day would be tolerable. A few weeks highly unusual.

The thing is, that base load is much smaller than the current base load provided by nuclear + legacy fossil fuel plants. Germany has been shutting them down by the GW. Nuclear is nearly done. Coal is next unless they have to get out of Russian gas first. They're putting 200 billion into fixing their grid in the next few years. Something, tells me that they'll manage without blackouts. Blackouts have not been a problem so far when they grew renewables from next to nothing to the majority of their supply. Same for other countries across Europe.

A little bit of nuclear on the side will help. But it will mainly keep energy cost high because all that stuff has to be subsidized. Maybe these smaller reactors will reduce prices to be a bit less problematic. Otherwise, expect renewables to outgrow this stuff by orders of magnitude like it has for this exact reason. Add a few GW of nuclear over ten years or so. Add a few hundred GW of wind. Just wind. It might add up to 5%. That might be enough. I'm not against it. Just realistic about prices and amounts.

>The wind not blowing is a local effect. It's more rare offshore. It's very rare for it to be continent wide.

It also tends to line up with sunny weather when solar is pumping out energy. Today is very much one of those days. Solar is producing as much as nuclear in the UK right now for a fraction of the cost.

Every winter I see the same post about how little energy solar panels are producing today and why this means we need more nuclear plants.

Today the sun is out so it's all about how little the wind is blowing.

Northern Europe tends to have a few weeks in the winter where both wind and solar are problematic for a few weeks on end. Southern Europe is fine around that time and off shore wind is fine too.

That's why interconnecting grids is so important. We can have Norwegian hydro and Moroccan solar picking up the slack. Plugin all those nice new EVs with vehicle to grid capability built in (Ford, Kia, and a few other manufacturers produce these already) and we're talking a few GWH of standby battery capacity ready to balance the grids as well.

That 5% base load is what we need to secure long term. But given that we essentially have that right now, there's no urgency for that. Maybe it's 10%.

Before anyone asks, I'm basing my numbers on what Michael Liebreich cited in his podcast a while back. Interesting person. He's a british conservative with a technical and economic background. Not your usually hippy environmentalist. Founder of Bloomberg New Energy Finance; generally seems to be extremely solid on his numbers. The economics in this area are quite fascinating. If you have better numbers and sources, please share. Base load without numbers is just people waffling.

I've never actually seen numbers which take into account pumped storage, grid interconnects, actual weather, low-hanging-fruit demand shaping and the anticorrelation of wind and solar but I wish somebody would produce some.

There's a lot of very shallow analysis out there though that cheerleads some approach with some very trivially refuted assumptions.

My sense is that the price of all of those things combined is currently about 90% of the cost of nuclear power official and about 60-75% of the cost of nuclear power that pays for its own disaster insurance.

While my own model is most assuredly flawed, I havent really seen a good alternative.

It's been a while since I read it to know if the specifics you're after are in there, but I wouldn't be surprised if you could get a directionally-correct reading from https://www.withouthotair.com/. Provisos obviously include changes in costs since it was published, but it would give you a starting point.

> Every winter I see the same post about how little energy solar panels are producing today and why this means we need more nuclear plants.

The low-wind periods last December had a peak of 1.2 GW solar.

Combined wind and solar right now are producing only very slightly more than a tech that has had zero real investment in thirty years. I wouldn't brag about that.

Is the eye watering cost that inhibited investment for 20 years supposed to be bragworthy? It's not like they were ever banned. They just need ridiculously high subsidies to exist at all.

Unlike wind, solar and pumped storage.

No, it isn't. It's a colossal example of both complacency at big engineering firms, and wider myopia about climate change.

Nevertheless, the >30-year old nukes still delivered what every single technology you have listed categorically failed to deliver. I have provided ample, repeated evidence of that. They failed and fossils came to the rescue. Repeatedly.

Why should they get even half the subsidy nukes get (which is what they do get), when their output is simply useless.

I've posted numbers. Please, you want to show me storage can be economically done, today, do so. Or admit you can't because it can't. We no longer have time for utopianism.

Subsidizing renewables produces permanent reductions in costs, resulting in net public savings. Nukes have been subsidized forever, and costs have gone only upward.

There's no savings if there's NO POWER.

150 MW currently.

>Nevertheless, the >30-year old nukes still delivered what every single technology you have listed categorically failed to deliver.

This is simply, provably false. They have delivered cheap and green electricity and weaned us off fossil fuels gradually.

Theyre growing very very fast because theyre economic and a good investment. Nuclear is not because it is not.

Throw $4 billion of subsidies at solar, wind and pumped storage and youll get more power when you need it than if you did the same for a nuclear power station.

> They have delivered cheap and green electricity and weaned us off fossil fuels gradually.

I'm sorry we're just living in fantasy land now.

We're not weaned off fossils. Not even close. That's just a lie.

In the UK solar and wind went from < 5% to 30% in the last 7 years because it's cheap to build. Nuclear's growth has been zero because it's ridiculously expensive.

180 MW today.

https://grid.iamkate.com/ says 4.5 GW or 13% today.

Where do you get that data from? I checked on electicityMap [1] and the British wind farms are producing more than 6 GW at the moment [2].

[1] https://app.electricitymap.org/zone/GB

[2] https://i.imgur.com/URVfMDT.png

https://gridwatch.org.uk/

currently 0.63 GW.

That's interesting that they disagree. Gridwatch has been around a while but I'll have a look.

edit: yep, electricitymap's free figures are not even close to live.

It looks like electricitymap is showing "estimated" data for the UK. It's fairly calm at the moment, and it probably doesn't take weather into account when estimating.

I think "estimated" is a recent addition they show, when the input data sources fail. Which unfortunately they do quite often. But then, I am not paying for the service, so who am I to complain? :)

When values are non-estimated, they are about 2 hours behind in the free version.

Could also be called "fraudulent" data

Not so. Fraud requires an intent to deceive to make unlawful (or unfair) gains. They're probably just abiding by the license of the data or struggling with a technical limitation.

lol - fraud isn't purely limited to gains. Where the heck did you get that from? Fraud comes into play any time someone intentionally tries to mislead - for whatever reason. And there has been plenty of fraud around energy reporting - again, for various reasons.

> fraud isn't purely limited to gains. Where the heck did you get that from?

The common and legal definition of "fraud".

Yep, this. Fraud is not a synonym of "lie" or "wrong".

Someone is gaining from wind farms claiming power output numbers they don't produce. This website is grossly overestimating the output of wind farms. Seems plausible there's a relationship between those.

It seems the technical problems are resolved and currently the data is no longer "estimated". 3% of the consumption comes from wind, and 17% from solar, which isn't to bad. While the UK certainly isn't the optimal place for solar power, it shows that it can deliver a large contribution. Especially as there is still so much roof area to use that it should be a no-brainer.

Yeah, and the UK's wind capacity factor should be better than average because most of the recently-added capacity is offshore, which is more expensive per GWh produced but produces it more reliably. There just doesn't seem to be a realistic non-nuclear path to ending fossil fuel dependence, and green groups have got away with obfuscating this fact because most of the press supports them.

> Please, show me an estimate for an alternative fossil-fuel system that can guarantee that shortfall.

I like nuclear. I have always liked nuclear. However: it's always sunny somewhere on Earth, and the total amount of solar energy delivered to us by nature's finest fusion reactor is something like 170TW.

If you will indulge me the geopolitical fantasy of putting a suitable planetary-scale transmission system in place, is there any particular reason we couldn't switch to all-solar?

Reason why is as you state:

>geopolitical fantasy of putting a suitable planetary-scale transmission system in place

That combined with geo-politics, war, big oil etc.

Yet we do manage to wire the internet accross oceans and continents.

The infrastructure scale is not even in the same ballpark, or even neighbourhood.

Besides the fact that you can't just "make" internet from a coal/petrol/etc generator somewhere nearby.

Imagine a ship's anchor damaging a cable carrying a couple megawatts of power (this happens frequently to the undersea fibre optic cables, and even those are a nightmare to repair).

Internet connection doesn't suffer as much power loss over larger distances

The internet does not go from A -> B it is a web. You can (and some countries do) cut yourself out of it. But you cannot cut others out of it.

Solar generation of power across countries will pretty much go one way. The most obvious and likely one is from North Africa to Europe. If the North African states decide to turn off the power to Europe they can. This is why geo-politically you cannot rely on this as a 100% solution. It can add to a solution, but it is not a good idea to let another country hang a sword over your head like that. (Or more precisely to hang the sword there yourself)

> any particular reason

Indulging such imagination, which is valuable, one particular reason, as you probably realize, could be resource limits when it comes to actually scaling up solar generation and transmission facilities.

Plus, electrification of transportation, if beyond electrified rail, still calls for storage. Here's an analysis of material inputs, focussing on vehicle electrification:

https://www.minersoc.org/wp-content/uploads/2019/05/3ICM-Mic...

Europe has its winter all at the same time. Everyone's making their morning tea and heating their houses at the same time.

We could, maybe, import all that energy from solar farms in North Africa, but (a) that requires a willing counterparty, and (b) is that actually technically feasible? As in, sign here and pay, we build it for you.

"(b) is that actually technically feasible"

It is, see the Desertec initiative.

https://en.m.wikipedia.org/wiki/Desertec

It would be just a huge investment, which is why it did not get very far, so far.

Also there is indeed the political factor. North africa is not so stable, but of course willing for investment.

But even with solar in southern europe alone, you would get quite far, combined with wind and hydro. And if you have excess in summer, you could store that in various ways for the winter.

In other words, I believe we do need huge investments into the grid now, connecting europes south and north. And then companys can build more of all kinds of solar and wind farms.

Correct me if I’m wrong, b it we can maybe, just about, for a lot of money, store electricity for a couple of hours. Maybe days with hydro. Multiple months seems like an impossibility.

The link you provided seems to just mention plans and studies, not building subsea electricity cables. Also says

> The project failed twice due to the problem of transportation and cost-inefficiency.

Like I said, it would be a huge investment. But without huge investments, we will not get rid of the dependency of oil or dictators either way. The question is what kind of investment. And the studies of desertec implied that it is technical possible. Since then solar dropped in price a lot and oil the other way up.

And if you have a european wide grid (high voltage, direct current) you would not need to store for months, because the sun is still shining in the south and wind is blowing somewhere. You only would need to store for the rare situations of little wind and little sun.

It will be a mix either way, I am not against nuclear as base load, or backup. But they are expensive, too.

At least nuclear is more under your control geographically. If an enemy missile can hit a solar field in a North African country and cripple you, you aren't very resilient.

How about an enemy missile hit your nuclear plant?

Did you understand what I was saying?

Not sure. First I thought you did sarcasm, but it seems, you are actually serious.

Well then, when you have lots of solar and wind fields, then it doesn't matter too much, if one gets destroyed. But even those you can defend, even with your own troops, by making adequate contracts.

On the other hand, if you have to rely on nuclear - then a potentiell enemy only needs to target those and you are screwed in many ways. It would be not too hard, for a dedicated commando to blow up, or at least render useless, a nuclear reactor with enough preparation.

But when you are only thinking in military terms: the hard parts to defend with dessert power are the power cabels. But they are also an essential part of nuclear plants and you cannot defend the whole grid.

Er no, wasn't being sarcastic. My point was that importing solar power from North Africa means you're less in control in the event of a conflict.

Your point about cabling is equally valid.

Your point about nuclear less so. I'm more comparing locally generated power, where your power generation can be covered by your own defences, with remotely generated power, which can't.

" But even those you can defend, even with your own troops, by making adequate contracts."

Making a deal with the moroccean government, would include their troops defending the investment, or could be modified to include own troops.

But like I also said above, the south of europe has plenty of sun, too.

If this is priced into the cost of the solar, then fair enough, although to bring this particular risk down to what it would be if the generation were in one's own borders would I think be prohibitively expensive, or impossible.

Little wind and little sun is every calm night. It is definitely not a rare occurrence.

But nights don't last for months (except in the far north, but they have plenty of hydro).

Storage only for nights, is very managable.

Can't think if any good reason we should.

Actual deaths from solar power are higher than actual deaths from nuclear energy, per GW delivered. Solar death is mostly from falling from high places, but so what? Death is death.

There's a conversation to have about the long tail but I'll probably start calling people cowards at some point so lets not.

It's easy to count somebody who fell off a roof. Statistical deaths from cancers are much easier to sweep under the carpet and most pro nuclear propaganda death counts I've seen do precisely that.

Are you alleging a conspiracy? Show me your numbers if you want me to take you seriously.

https://www.i-sis.org.uk/Chernobyl_Deaths_Top_a_Million.php

Im not sure if conspiracy is the right word. The special interest groups bullshitting everyone are not necessarily coordinating. Is religion also a conspiracy?

Never mind.

I believe that I'd feel safer with nuclear if they paid for their own damn insurance and raised the liability cap to "unlimited" but apparently they dont feel its safe enough just yet to raise their financial exposure. What if fukushima happened again? Dealing with that was expensive!

Perhaps you could explain why they cant volunteer to shoulder their own risks if theyre so safe?

In SPITE of free insurance they still need lavish taxpayer subsidies to be financially viable while wind and solar dont.

> if they paid for their own damn insurance

...including a policy to pay the cost of decommissioning.

There isn't a production nuclear power plant in the world that has ever been fully decommissioned. The earliest production plants in the UK have been taken out of service, but decommissioning (so the land they were built on can be re-used) is expected to take the rest of this century.

Then everything else would need a similar policy for fairness. Coal plants and their ash ponds are notorious.

I don't disagree. Corporations leaving environmental cleanup as a problem for government (i.e. taxpayers) and future generations is a severe form of externalising costs, which I think amounts to theft.

I've not seen any numbers produced in the last decade that agree with your claim, so maybe you should provide your numbers.

https://ourworldindata.org/safest-sources-of-energy

Are you saying more people have died from falling while installing/servicing solar than have died due to Chernobyl and Fukushima?

Per GW of delivered power. Yes, that's correct.

There were 1 person that died in Fukushima. Some more, mostly elderly people, died during evacuation/displacement.

In Chernobyl around 50 people died as direct consequence of the disaster. As far as I know there's no studies showing increased risk of cancer in the population that lived in and around Chernobyl at the time. If Chernobyl showed us one thing it is that radiation is not as dangerous as we believed.

There were sharp increases in some cancers, however from very small bases and of cancers with very low mortality, as far as I recall.

I think it's worth assuming some reduction in life expectancy for some people, but notably the estimates of lost years of life have dropped year by year, and will likely end up being negligible.

I've said it before: we could have a Chernobyl every year, and it'd still be worth it (reducing deaths) if it got us off coal sooner...

As far as I recall, there were some measurable increases in cancer incidence among children in Belarus, mostly due to milk intake

However, all that was completely drowned by a collapse of life expectancy resulting from the chaos of the 90s. We're horrified at a few tens of extra cancers, but 100M men losing 10-15 years of their life due to socio-economic circumstances doesn't raise anyone's eyebrow

Long term Chernobyl numbers[0] are highly contested.

[0] https://en.wikipedia.org/wiki/Deaths_due_to_the_Chernobyl_di...

Can you cite sources for your claims?

I am personally doubtful of your comments around deaths being more frequent. Statistics around the specific cause of death would also be appreciated.

>Actual deaths from solar power are higher than actual deaths from nuclear energy, per GW delivered. Solar death is mostly from falling from high places, but so what? Death is death.

This is not a serious argument. Large scale solar farms are hardly more than 6 feet off the ground and would essentially eliminate this risk. I expect the number of deaths and decrease in quality of life attributable to fossil fuels is far far higher to speak nothing of Climate risks in general.

> If you will indulge me the geopolitical fantasy of putting a suitable planetary-scale transmission system in place, is there any particular reason we couldn't switch to all-solar?

Just how much would such infrastructure cost? How much resources would we have to mine to facilitate such a massive overhaul to a non-centralized grid?

I think when renewables are presented as the only answer it is underestimated how much it would cost in investment in storage and the grid, to a point that I feel nuclear is a fine solution for base-load.

I have no numbers though so I cannot back this up or anything.

It reminds me of non-developers with ideas for apps/software.

As if the idea of getting all our energy from the Sun is an original thought.

How do we accomplish this?

"I don't know, I am just the idea man. You do all the engineering for my idea and we will split the profit 50/50!"

Can you check your numbers. 170TW sounds on the low side, PW would be more like it.

You're right: ~170PW is the correct number.

https://news.mit.edu/2011/energy-scale-part3-1026

Both nuclear and a planetary grid are theoretically possible. The real question is what's quicker and cheaper. It seems very unlikely that a planetary grid would be the winner here.

Interconnection is a liability.

Transmission losses are something like 3-10% per 1000km for overland cables, and 60% for underwater cables. The Earth has 12000km at the equator, so solar power at night would have to be delivered at 18-60% losses to some areas in the best case; much worse in practice, since you can't realistically have overland cables connecting Eurasia to the Americas.

You also are unlikely to have enough space and enough raw materials to construct solar panels that can provide the entire earth's energy needs all-day round, given the huge transmission losses.

Furthermore, solar panels have a very short life span - barely 10 years for a decent yield - which means that you would have to constantly mine hevay metals to build new solar panels at a rapid pace just to maintain this setup.

It's perhaps a possibility for a distant future, but definitely not something that can be contemplated as a serious solution to the 0-emissions world that we need to hit in the next 30 years.

I’m not an expert in this field at all, but from the bit of research I did to try to verify these numbers, the estimates presented here about solar longevity and long-distance transmission seem unrealistically pessimistic.

Losses for 800kV HVDC systems are typically documented at ~3.5%/1000km or ~5%/2000km[0], and state-of-the-art 1.1MV systems have losses of ~1.5%/1000km[1][2]. Lower voltage systems have higher losses but I’m not sure why a lower-voltage system or AC system would be used in a project like this.

I can’t find any information about increased losses in submarine cables at all; where is that information from, and what is the mechanism that causes that added loss?

For solar panels, as far as I can tell, capacity loss occurs at a rate of ~0.3-1% per year depending upon the type of panel, with a median of 0.5%[3]. More modern panels have lower losses than older panels, presumably due to manufacturing improvements. So, after 10 years, you’re talking about something that will produce—on average—5% less power than it did when it was brand new. That’s not something that is broken and needs to be replaced the way you seem to be suggesting.

[0] https://publications.jrc.ec.europa.eu/repository/bitstream/J...

[1] https://news.ycombinator.com/item?id=29232512

[2] http://en.people.cn/n3/2018/0622/c90000-9474097.html

[3] https://www.nrel.gov/docs/fy12osti/51664.pdf

If you have enough installed capacity that the production averaged over a year is sufficient, you can store excess energy as hydrogen or methane and turn it back into electricity while renewable production is insufficient. The amount of storage you need is a function of how much capacity you install, how large the grid is, and how smart e.g. electric car chargers are. Some clever economists probably can figure out the cheapest configuration.

Currently it's much cheaper to just burn fossil fuels or import power from your neighbors than to build hydrolyzers, which is why you don't see many, but pilot facilities in the multiple-MW range exist, so the technology is available.

conversion has losses - if it was remotely economically viable people would be doing it ALL OVER the place. Energy storage is the magic bullet to make renewables truly feasible for base load support and if this was remotely viable AT ALL someone would have implemented it already.

Just because a tech is available doesn't mean it's practical for every day use.

Conversion has losses, true. For electricity to gas and back the losses are around 50%. That's perfectly fine for a technology you need for maybe a couple of weeks per year. The real question is how expensive the equipment is. Power is essentially free for renewables, all costs are depreciation of equipment.

And as I said above, people _have_ implemented it already. There are dozens of MW of hydrolyzers installed in Germany alone.

It's also just turned warm and sunny in the UK. I turned the heating off for the first time in months today and solar is producing just as much as nuclear right now.

At a much much lower cost.

Those wind shortfalls have a strong tendency to line up with good weather when solar produces more and heating is less needed.

The UK also has more pumped storage coming online and octopus even has home electricity tarriffs that promote time shifting demand.

I think "plans" is putting it quite strongly re the UK planning to move to electricity for heating. We've had a "plan" do something about emissions in <today> + 5 Years for about 40 years now. So I wouldn't rely on anything actually happening. Especially given that it would be practically impossible to switch from gas...

You should see how little those solar panels generate at night!

While you're not wrong to imply PV doesn't work at night, concentrated solar power (CSP) actually does. It achieves this by heating salt into a molten state which can then discharge heat (and drive turbines) overnight. CSP is very promising for helping manage peak loads in place of gas, too.

There's also batteries: virtual power plants that give us the ability to arbitrage between when power is cheap and when power is needed. These can be rolled out in conjunction with residential PV systems and eventually pay for themselves. A similar approach can be used to store hot water produced during the day by PVT (photovoltaic-thermal) cells.

See:

https://en.m.wikipedia.org/wiki/Thermal_energy_storage

https://www.energy.gov/sites/prod/files/2014/01/f7/csp_revie...

https://en.m.wikipedia.org/wiki/Virtual_power_plant

CSP plants have never gotten beyond the prototype phase because no one has figured out how to make them economically viable. One near me in California just shut down because maintenance costs were skyrocketing.

Batteries, in their current form, have all sorts of issues that haven't made them economically viable either. Liquid electrolytes (liquid batteries) do show promise for grid level storage, and I am pleased to see lots of activity around them but we are still probably 5-10 years away from anything worth scaling up around.

Hot water is fine for extreme micro uses - individual buildings - but again not even remotely viable for something at scale, let alone replacing base load generation for the grid.

Base load is the real nut to crack - and so far solar is no where near being anything but a nice supplement for base load.

I'll not argue with anything in your post and accept the issues with CSP as per the status quo, but hot water needn't be a micro, single household scale. I've benefited from a CHP plant powered district heating system and it more than halved my energy use. There's definitely nothing wrong with big centralised heat generation and storage schemes. It also helps address baseload.

The only good battery technology we have today - LiIon - relies on metal of which there just isn't enough on Earth for anywhere close to the required capacity.

Indeed, costs for storage are falling even faster now than solar or wind ever did. Solar thermal, already cheaper than nukes, is far from the best.

Reply on news.ycombinator.com