Adding capacity to the electricity grid is not a simple task | Hacker News Reader

Adding capacity to the electricity grid is not a simple task

economist.com

142 pointsmfiguiere203 comments

At my current company we're working on one of the problems involved in this in much of the developed world. Actually coming to the point where you can lay power lines is a huge issue that has to do with figuring out property ownership, the stuff that's already in the ground, water ways, roads, trees and more. People work months on finding a route that's actually possible and that sometimes involves paying a lot to property owners for using their land.

Right now we're creating a tool that gives the engineers insight into all these factors and automatically creates possible routes through all this madness. It's quite interesting to work on and fascinating to learn about all these complexities. It also doesn't scale well, we're building this for the Netherlands, which has a lot of centralized, (mostly) open data that can be used. In other countries that's not so much true.

Why shouldn't the property owners be paid for a profit taking enterprise to functionally steal their land? I understand that it has to be done to make the world work but it absolutely disproportionally effects rural people while the benefits go to the urban areas. This is one of those situations where a percentage of those gains flow back into the rural areas.

I don’t think he said they shouldn’t get paid.

He is just saying that sometimes the best solution/route is over private property where the owner gets paid.

Correct. It's also not just about the money, it also takes a lot of time and adds a large amount of uncertainty to a project. Not every owner is willing to comply, causing an entire route to have to be reconsidered, which takes a lot of time. Eminent domain is pretty much never used, because judges (in my country) will pretty much always side with the property owner, unless the utility can proof that there's really no other way (which they almost never can).

So on a 10km+ route with hundreds of properties, multiple municipalities or other public organizations and other utilities, this gets really complicated really quickly.

Eminent domain being “pretty much never used” is a bold face lie in the US.

https://www.usatoday.com/story/news/local/2023/04/04/bullitt...

Here’s a story from this week where a gas company took land from a forest to build a pipeline that the previous and current landowners had no interest in selling to the gas company.

I'm not in the US, hence me explicitly writing 'in my country', because I'm aware that more countries exist and that things aren't the same everywhere.

That's so bizarre. In my country only the government can force you to sell them your land under eminent domain.

Insane that in the US, companies can use eminent domain as well.

Much of the activity stems from a disastrous Supreme Court ruling a couple decades ago. https://en.m.wikipedia.org/wiki/Kelo_v._City_of_New_London

A lot of farmland and lovely homes have been unprotected since. Many Americans wish for this case to be revisited (if they know about the case.)

While Kelo was IMO a bad decision, it's also not the case that eminent domain is routinely exercised at low cost and effort in the US. Private property owners routinely block many types of development on their land and adjoining land. People like to decry NIMBYism all the time. I suspect building an interstate highway system in the US today would be effectively impossible.

ADDED: And, in general, exercising eminent domain should be hard. One can simultaneously believe it should exist and have a lot of safeguards against exercising it.

Eminent domain for a public road is one thing. Forcibly taking land for a private company to make profits on is entirely different and not comparable. That is the part I object to the most.

> I suspect building an interstate highway system in the US today would be effectively impossible.

The government should be able to take whatever land it needs for a highway, railway, etc. When we're talking about public infrastructure, the needs of the many outweigh the needs of the one landowner.

Perhaps the reward isn't enough if it is that difficult to get permission. Landowners near me compete for cell tower leases because it is absolutely worth it. I suspect profit sharing on the electricity sold through the line would grease the wheels a bit.

Perhaps landowners have unreasonable negotiating leverage when a small number who have little to gain or lose can stop a project that is worth billions to society as a whole.

If you need to get permission of thousands of people, and any single one can hold up the entire endeavor, then the economically rational thing to do is to be the last holdout, as you will get paid a lot more. Everybody loses in that scenario except the holdout.

Negotiations can be done en bloc but the downside is that it may take decades to do anything big.

In the US, class action lawsuits allow multiple entities to be rolled into a class to sue another party. Is there an equivalent in the other direction (i.e., could a utility company roll everyone into class)? I see lots of possible problems (i.e., class participation allows opt-out).

The dynamic of class action lawsuits isn't about the quantity of class members, but rather that they're unenumerable. As such they don't really make sense to apply to defendants, and would have terrible results if they were.

A plain old suit with many enumerated defendants would work for this topic though, assuming there was a cause of action.

That is exactly my point. Everyone gains massively except the poor guy who has to live with the problem. You don't see something terribly wrong with dumping your problems on random people just because they happen to live out of your sight? If society gains billions then surely you can spare some percentage of that gain to lift up the rural areas and people. Instead the coasts look at us like rubes and treat us with distain while we feed them and get our land forcibly taken. It seems like such an attitude would eventually cause a political rift in such a society.

The problem isn't that the little guy has too little say under the current system, but far too much.

That cost comes out of the pocket of everyone else who has to pay more for electricity. As such it isn't good for society to allow high prices.

Land owners do need compensation, but we need to be careful to not let that compensation be any higher than it must be.

Is it really not good for society to allow high prices? A high price on electricity would lead to consideration of alternate solutions, such as just not doing the thing which requires it or local power generation. The problem we have is that negative externalities aren't priced into fossil fuel alternatives

An advantage of increased awkwardness in dealing with local landowners would be that local organisations are incentivised to arrange a solution voluntarily, strengthening the local community, such as community owned and administered infrastructure that was constructed without government having to use eminent domain

They pay one time pennies for land and run billions of dollars of value across it. It is insulting.

Power companies have done this in the past; there still exist farms that have perpetual free power from an agreement back in the rural electrification days; we get a right-of-way for the power lines, you get free power.

It's often much easier for the power companies to just negotiate with the city/state to use the already existing road right of ways, even if that's longer.

In countries that do not have a cadastral property register, it may not be simple to go from "GPS location" to "mailing address to offer to pay the owner for access to their land"; finding that can be a substantial cost in itself.

It's a multi-player prisoner's dilemma. Say you need the co-operation of 1000 landowners.

- Everybody co-operates, everybody gets paid a little bit.

- One or two people make unreasonable demands, the builder caves and they get paid a lot.

- Several people make unreasonable demands, nobody gets paid.

It's more complicated than that because multiple routes are always considered simultaneously, but the basic large number non-iterated prisoners dilemma problem applies.

>unreasonable

I think therein is the issue. Who is to say that the small piece of property isn't worth the "unreasonable" amount to that owner?

My home is worth $X on the market, but I don't want to sell it. Why shouldn't it be worth 10 * $x if a company wants to buy it from me? Why should I be forced to sell?

It's the time-old public/private issue; with money it's easier because it's fungible. Eminent domain has existed for millennia; this is just an aspect of it.

In general the common result seems to be that for the "public good" the government has relatively large leeway (+/- political pressure) for things like roads, freeways, transit, utilities, etc.

The line gets blurred when some of those are technically for-profit companies like some (many) utilities are. Often the end result is relatively fair for "everyone" but still the landowner is most affected (things like requiring power lines to be buried, or the power company has to build a driveway for the land owner, etc can happen). At extremes you have "town eminent domains your house so Walmart can build a parking lot on top of it."

Some cities plan very far in advance and basically put a "we will buy this property when the landowner dies" lien on it; and wait out the 30-50 years before doing the project.

For reference, here's how we used to do things:

https://en.wikipedia.org/wiki/Cataloochee_(Great_Smoky_Mount...

>In the 1920s, the increase in tourism and the destructive effects of logging gave rise to a movement to create a national park in the Smokies. For this to happen, the residents living in the proposed park boundaries would either have to sell their land or be forced out via eminent domain. In 1928, Reverend Pat Davis broke the news to the residents of Cataloochee at Palmer's Chapel, telling them the valley was within proposed park boundaries and that they would be forced to sell. Hattie Caldwell Davis, who was at the meeting, recalled women crying and men threatening to dynamite the roads and shoot anyone who tried to enter. Some men had the idea of blocking Cove Creek Gap so the government representatives would have to come in from the Tennessee side, where Mount Sterling residents would surely stop them.

It’s not unreasonable for extremely small portions of private land to be used for infrastructure, in the exact same way that easement rights are not unreasonable. You can’t buy all the land around my lot and tell me you value right-of-way on your property at $1 billion.

This is why the government has eminent domain authority. Either take a price that you feel is fair, or hold out and let the government decide what price is fair.

Eminent domain for a public use is one thing, stealing private land and giving it to a profit taking company just because you can is another.

Using eminent domain to run utilities is a quintessential use case for eminent domain. The owner doesn't necessarily need to be forced to sell the land, they could be forced to grant an easement instead. A utility easement is something almost every land-owning American has to deal with.

An easement that forever stops me from planting a tree or building a building is not simple ROW. It is a takeover of my property.

It's not exclusively your property, because it derives a large portion of its value from existing as part of a community.

A community that requires many shared goods: water, power, transit, sewage, waste disposal, etc.

Each of which require sacrifices to be made by all so that they can functionally exist.

Absolute ownership = no shared infrastructure can be constructed = everyone's land is worth less

They require sacrifices to be made by those who preserved land. I realize that we cannot have absolute ownership. What I see in practice is that farms and open spaces are run over roughshod. Road planners try to avoid tearing through businesses and neighborhoods while gleefully destroying a century old farm. It isnt a shared infrastructure or sacrifice and you know it.

A fair system would be blind to what is on the ground and build the lines or roads without regard to who owns it or what is in the way. Otherwise the poor and rural will always be looked at as cheap useless fools to be taken advantage of.

If you've got a bone to pick with planners, then you're drawing the wrong distinction.

They don't care about rural/farm vs urban/business.

They care about tax revenue and voters.

It's always going to be an uphill battle making the argument that $ tax revenue / acre farmland is more important than $$$ tax revenue / acre sub/urban use. And due to density that farmland represents one family of voters vs the >1 an equivalent urban footprint would impact.

The fact is that the rural US has benefitted from the US's very robust freight rail network, almost all of which was either taken via eminent domain or before the land was parceled out.

If we tried to rebuild the train network we have today using the protracted legal process we have for reallocating land, it would be 100% impossible.

The rail network is a big part of the reason we're in this mess, though.

The "checkerboard" system granted alternating squares of land to the railroad companies as an incentive to build out the network. The railroads sold those squares to various entities.

The result is that much of the public land in the Western United States is "corner-locked". There are two parcels, diagonally adjacent, but a new easement would be required to cross between them.

Even if you could get the corner easement you'd be running lines that zig-zag all over the place, so now the private owner can extract whatever the savings are for a straight line.

https://www.onxmaps.com/onx-access-initiatives/corner-crossi...

This seems like a completely unrelated implementation detail.

Seriously. The government could have just leased rights to the land for 100 years to the railroad companies.

We don't need to go full PRC, we can just make land grants that make geometric sense.

If the concern is getting multiple companies to cooperate and preventing them from accumulating too much land/power, leases are a good solution.

Land leases are not bad things just because the PRC happens to use them. The USG leases a crapload of land to companies for various profitable enterprises. It's a tool in the toolbox.

> I understand that it has to be done to make the world work but it absolutely disproportionally effects rural people while the benefits go to the urban areas.

This seems like a bit of a stretch. The usage of extremely tiny portions of rural land for infrastructure pales in comparison to the ludicrously large subsidies of rural areas by urban areas.

> ludicrously large subsidies of rural areas by urban areas

Which subsidies are you referring to? The department of agriculture does have subsidies, but they measure in the (<10) billions which I don’t think we can call ludicrously large in comparison.

I'm mostly just talking about tax revenue versus public spending, since that's probably the easiest to get objective numbers for. But of course it's still going to depend on where you draw the line between "urban" and "rural" (I would probably want to look at dense urban centers versus everything else). With a quick Google you can find a bunch of studies showing that urban areas subsidize rural areas as well as a bunch of articles "debunking" those studies, so I'm sure anyone can find the conclusion they were looking for.

There is also subtler stuff that isn't exactly "subsidy" but is perhaps suggestive of societal conflict between urban and rural areas, for example vastly disproportionate political representation (in the United States).

They should absolutely get paid (and they are in most civilized countries). But they shouldn't have a veto right. I private ownership of land is a weird concept anyway.

They get paid a pittance compared to the value of what is provided.

In from the Netherlands too,can you find me the company name? I think someone is interested in this as well.

I used to be a permitting attorney. My wife is presently a prominent permitting attorney who works on linear transmission corridor projects at a large firm. Unless we change the permitting process, and several federal environmental laws in particular, we will never realize even a fraction of the capacity required to hit our renewables goals. We can’t get a single new line across the cascades in Oregon, much less the dozens that will be required. Tidal energy faces the same issues vis-à-vis the load centers on the other side of the coast range.

There will probably be a lot of pressure to change the permitting process in the near future. Power outages, along with rising prices, are one of the events that most reliably get politicians thrown out of office.

They'll just blame the 'greedy' energy companies for societies woes.

We are facing a major crisis in South Africa where we used to have more than enough power, 30 years ago, but the government neglected the maintenance of the network, and now we cannot make enough electricity, nor even maintain the existing levels!

It's happening elsewhere too; in NL, there's plenty of incentives to go green - cheap, subsidized solar, hard push from the EU to transition to electric vehicles in less than a decade , etc - but the grid capacity just isn't able to keep up, even with multi-billion investments, which results in some businesses not getting a connection to the grid or new green energy projects being delayed.

Which is more an inconvenience than a problem for the general population, but still.

The other thing that's happening is that while the energy companies, with help / subsidies from the local and EU government, build large offshore wind parks and the like, at the same time the capacity is bought up by large new datacenters from the likes of Microsoft and Google, meaning that they don't actually replace grey energy sources. Said datacenters also use up drinking water for cooling, for some reason.

super-wealthy polluters claiming that it is "too hard" to go green ?

> capacity is bought up by large new datacenters from the likes of Microsoft and Google

That is surprising but also makes a lot of sense. Companies go for the already built cheaper option to maintain that they’re going green while taking it from others! That’s actually kind of harmful and they should have to pay for building energy if they want the moral credit for it.

Didn't someone actually attempt to poison the manager that was asked to fix the problem?

Yes that was the last CEO Andre de Ruyter. That’s just the tip of the iceberg though.

The level of dysfunction where an electricity guy needs tight personal security because he wants the power stations to actually run is just breathlessly crazy.

Why would someone even poison him for that…They don’t like stable electricity or what?

He identified and thwarted a lot of corruption [0]. So it’s more that people don’t like their little cabals and mafias being disturbed in the process of creating stable electricity.

[0] https://youtu.be/6gF2dYQ-NdM

South Africa is the slow boiling frog.

Functioning electricity network + DEI = South Africa.

Every black-majority country is like this…incompetence and corruption takes over. I always wonder what’s wrong with my race.

> Every black-majority country is like this…incompetence and corruption takes over.

I don't think that's true. For instance, São Tomé and Príncipe is well run by all accounts I've read. I'm sure there are others, that's just one that comes to mind right now.

Population below poverty line - 66.7%; Unemployment - 15.9%, according to the World FactBook.

Does not strike me as a well run country despite their little population.

1- https://www.cia.gov/the-world-factbook/countries/sao-tome-an...

It's a small archipelago nation with a tiny workforce and few natural resources, so they're more or less destined to be poor. I don't think that reflects poorly on the quality of their governance though.

They've had a functional multi-party democracy for decades (a few small failed coups, but with numerous peaceful transitions of power between parties and no civil wars.) Their primary education system is pretty good and literacy rates are high. Perception of corruption is average and trending down. They rate well for free speech, political and economic freedom.

> "At the turn of the century a couple of power plants a year might be connected ... new plants often using the same connections as old ones."

It's still generally true that the old grid infrastructure is re-used.

The UK has many closed coal, oil, gas, and nuclear power plants on or near the coast. When those plants are decommissioned or demolished, the grid infrastructure that was built for them (substations, transmission lines) is usually left intact.

New off-shore wind farms can now use those access points, which greatly reduces the cost of connection compared to having to build everything from scratch.

In theory yes, but the locations in the UK don’t coincide.

The old coal plants were built nearer to consumption, most of which is in the South East of England. The planned wind farms are mostly in the North Sea, far from South East of England. It will cost billions to build the infra to connect the north with the south. It’s not a simple matter of reusing what exists.

OK, but you left out the existing large wind farms, most of which were built right next to the south east of England, and presumably used the connections that OP was talking about. https://en.wikipedia.org/wiki/List_of_offshore_wind_farms_in...

£50 per UK resident provides several billion. If it’s important enough, a few billion over the next 5 years to secure electricity needs isn’t actually a hurdle.

You can't address every problem like that. Those £50 a time would quickly add up. Also that's several hundred pounds per household right out of discretionary income.

Agreed, which is why the notion of important-enough comes in.

I think that stable and sustainable utility supply would fall quite high on the priority list for most households.

Oh sure, we have to do what's necessary by definition. It's another extra cost falling on households at a time when costs for everything are piling up.

Is there a RoI in terms of reduced energy costs for households?

As with everything, if it's actually important to the right people the money can be found. Businesses operating in Britain need reliable, affordable power as well. Putting the cost burden on retail customers is a political choice.

(and small businesses; the recent price hikes have been far higher than a mere £50 and have been putting restaurants out of business)

Building proper interconnect between the north and south isn’t “one of many competing interests.” It needs to happen to stabilize the UK grid, everyone accepts that it needs to happen, it has a multi-decade payoff time. So just figure out how to get it done now and accept that it’s not going to get cheaper. Then move on to things that are less critical and start pinching pennies there.

> everyone accepts that it needs to happen

This is the UK. Just because something is obvious doesn't mean that conservatives and NIMBYs are going to accept it until they, personally, are sitting in the dark.

And at that point, they'll blame the previous government ;)

The North-South interconnects are already happening. The Western HVDC became operational in 2019[1], and two separate Eastern HVDC links (Eastern Green Link 1 & 2) are pretty far along in the planning process, with EGL1 expected to begin construction this year and EGL2 in 2025[2].

[1] https://en.wikipedia.org/wiki/Western_HVDC_Link

[2] https://en.wikipedia.org/wiki/Eastern_HVDC

Almost all the UK nuclear plants and oil-fired plants were built on the coast. (The only nuclear plant ever built inland in the UK was Trawsfynydd in Snowdonia)

In the case of nuclear, this was to ensure reliable supply of cooling water. In the case of oil, it was to make it easy to unload crude directly off ships. The oil-fired plants were often co-located with oil refineries, many of which have now also closed.

There's a good 3 part series from Guy Martin on Channel4 recently. Would watch far more hours and depth of this content though.

https://www.channel4.com/programmes/guy-martins-great-britis...

anything by Guy Martin is great. He just seems an all-round wonderful person

A large proportion of the UK's coal power stations were clustered in the Trent Valley: https://en.wikipedia.org/wiki/Megawatt_Valley

Over the last few years, lots of renewable generation and storage projects have taken advantage of the grid capacity in this area, but all the low hanging fruit is gone now.

These days, it is not usual for projects to be given a connection date in the 2030s due to the requirement to reinforce the transmission network.

Which is a bit curious given how fast electrification originally happened.

I suppose the capacity of that infrastructure was much lower than the requirements nowadays, and labour was dramatically cheaper back then.

No, there was simply much less red tape.

And early in electrification the whole local grid going down for awhile because who knows why wasn't a major issue, it was almost expected.

Some towns didn't even have power until near nightfall as a planned runtime; why bother with power during the day?

Now we're used to 100% always-available power at all times, and the demands on it are growing. Things like power walls may become nearly free as the grid maintainers try to flatten demand so they can run the grid closer to capacity for longer each day.

Mainly due to NIMBYism and planning/environment laws becoming more complex.

E.g. to get a 132+ kV circuit constructed you need to go through the DCO/NSIP process which takes many years. https://en.wikipedia.org/wiki/Nationally_significant_infrast...

The CEGB constructed the 400 kV supergrid in the space of 15 years during the 60s and 70s. A project of such magnitude today would most likely be tied up in the planning consent process for 10+ years.

Mainly due to NIMBYism and planning/environment laws becoming more complex.

I agree, but also think that outside of the nimbys, other processes could be vastly improved.

There should not be 30 agencies involved, or even 5. There should be 1.

And that 1 agency can cover all of this, run all studies in parallel, and be a single contact point.

It should take days or weeks. Not months or years.

For NSIP projects, there's already a single government body (Planning Inspectorate) that is in charge of the process.

Agreed, the process needs to get a lot quicker for all infrastructure projects.

> It should take days or weeks. Not months or years.

I don't see how? A single court challenge + inevitable appeal would alone take up over a year.

It really happened gradually, though, and with much less surrounding complexity. To give an analogy, OSes developed really fast, but any significant change in the kernel now will require lots of caution and scrutiny.

Also huge advances can have lots of things not "solved" - the first vehicles sucked majorly and were finicky as all get out, but compared to a horse-drawn cart they were so amazing as to excuse all the failings.

Modern cars are so good that any advancement is going to be incremental, and so it will take longer.

HN may be interested in the rest of the special report, which is here https://www.economist.com/technology-quarterly/2023-04-08

Excellent series - one author, very broad and insightful.

Here in Canada my province went through a hurricane last Sept. I was without power for 10 days.

Much of the problems were due to poor maintenance; trees not trimmed, old poles not replaced, transformers old, salt spray on wires, etc. But the parent company crowed each year of profits and dividends. They also raised rates 5% to 10% each year claiming they had no money.

Solar would help but we also need a stable grid. But even with solar my local government doesn't even allow the use of battery storage in homes batteries can only go in garages.

I'm in Western Australia which like South Australia is a state with a very high solar | renewable input to our grid.

We're currently rolling out required remote management to remotely disable rooftop input on high solar input low demand days and likely to see a future where batteries are larger, isolated (for fire | emergancy), and serving local clusters of ~ 200 homes moreso than every home having batteries.

There's an efficiency and robustness at that cluster size.

I understand the need for grid stability, but just deciding "Nah, we'll turn off solar production en-masse" seems like it's not the best solution.

Long distance HVDC runs are apparently now so good that it is commercially feasible to drop a few million solar panels outside of Darwin, and run an under-sea HVDC cable to Singapore.

Why there's apparently no push to hook up WA to the rest of the national grid, I don't know. It would seem to be a great way to shift power from solar and wind between the east/west and help match up demand and production more.

> Long distance HVDC runs are apparently now so good that it is commercially feasible to drop a few million solar panels outside of Darwin, and run an under-sea HVDC cable to Singapore.

Whoa, that's impressive. Do you know what the per-mile cost would be? The great circle distance from Darwin, NT to Singapore is a little over 2000 miles; call it 2500 to account for routing around the intervening islands and such.

Though I'm not sure that Darwin is the best place to drop a few million solar panels, considering its stormy climate. A few hundred miles south towards the desert, maybe ;-)

The project I'm referring to is the Australia-Asia Power Link by Suncable[1].

It was supposed to be an AUD$30bn project, but I don't know the proper cost, but $30bn/4500km = $6.6k/meter. That includes the solar panels, and a massive battery too, as I understand it.

It went into administration in January, because the backers disagreed about further funding for the project.

Also, When I say "outside darwin" I didn't mean in the suburbs.

It may not be obvious just looking at satellite maps, but large areas of that part of Australia are subject to seasonal flooding. If you're building a project like this, you need to pick higher ground to avoid it being flooded.

They basically have two seasons: Wet, and Dry. During wet season, they still get plenty of sun.

Also, large areas of that part of the world are protected National Parks.

[1] https://en.wikipedia.org/wiki/Australia-Asia_Power_Link

It's more than a disagreement about funding, it's a disagreement about feasibility.

Long HVDC lines are feasible, as are long HDVC lines across (some) sea floors.

Even longer lines with intermittent stations across sea floors are feasible.

What's pushing the engineering limits and the stability of forward capital funding is a single really long undersea HVDC cable crossing multiple fault lines in one of the more volcanic earthquake prone regions of the world with a few very deep trenches.

There's no fallback in the quite probable case of disruption, no cheap way to add redundancy, and no cost effective way to deal with breakages.

The kinds of milestones being missed that are alluded to in the wikipedia article are feasible engineering solutions to route challenges.

Balancing power grids real time is a challenge, conventially excess production is handled by powering down sources (turning {off|down} {coal|gas} fired generators) or routing excess elsewhere (neighbouring countries|states).

Western Australia is remote .. there is no connected neighbouring grid (to NT or SA) and the distances are vast.

Winding down solar inputs from individual houses is a comparable way of shedding sources, even better is being able to balance the local cluster inputs between "main grid" (wider urban areas) and "local storage" (per N house battery banks).

If water wasn't such an issue here there'd maybe be plans to expand the main dam to handle pumped hydro (that requires both an excess of water and a lower elevation secondary capture dam) so ther are plans afoot looking at alternative short term storage solutions (in addition to battery farms).

> I understand the need for grid stability, but just deciding "Nah, we'll turn off solar production en-masse" seems like it's not the best solution.

It's not. But we've never before had a "we are generating electricity with a marginal cost of $0" situation before. As coal and gas generators had to pay for fuel they stopped generating rather than create an excess. But ... why turn off a solar panel?

We've come close with nuclear because it can't turn off quickly enough to match the normal demand cycle. As a consequence Japan has the most pumped storage per unit of generation in the world. They store their excess nuclear generation so they can sell the excess later. That's the best strategy for nuclear because it's so expensive, but in the renewable world its cheaper to over provision than it is to build storage (to some extent), so it's likely there will always be periods of excess.

So now we find ourselves in the novel situation of having literally GWh of electricity available for close to free. I can't imagine the situation remaining for that way for long, as we find a use for most concentrated waste streams. Here in Australia we even recycle our sewage solids. I presume that's the one reason for the current "green hydrogen" push in Australia. If green hydrogen doesn't work out, it will be something else like Aluminium or carbon free steel production.

Canada over regulating again..

> battery storage in homes batteries can only go in garages.

Batteries in that quantity are something of a fire risk, which is why some jurisdictions mandate they be "outside". Although this is probably just a different kind of over-caution.

True but right now there is a 1,000 litres of diesel in a tank in a room next to me.

Diesel is fairly safe because it is surprisingly difficult to ignite without pressure. Gasoline is a different animal.

And yet home heating oil tanks in basements are completely normal and under little regulation; tanks can and do fail, and if the spill is bad enough, it'll result in the house being "totaled."

The PV industry is saddled with an insane number of regulations designed to make PV installations as expensive and ugly as possible to discourage it. The utilities absolutely hate the idea of becoming a glorified standby network, but they fear people going "huh, I...don't actually need you at all" even more.

That's why when you see a typical PV installation, there's giant orange warning labels all over everything, multiple panels/junction boxes, etc.

It's also why you can meet all the state regulations, but still fail one of the multiple inspections utilities require before you can fully connect the system.

Same reason the utilities are pushing for state laws that automatically condemn a property if there isn't a grid connection.

How would the government even know you had batteries in the house? Build a suitable setup and tell no one

An electrical inspector would need to sign off on it and any connections to the breaker panel. An electrician would have to do that I don't have the qualifications to do that. I do have electronics technician certificate but not as an actual electrician.

If I did it on my own then I'd risk having my home insurance cancelled.

Just as with nuclear power reactors, I hope that we can develop standard transformer modules that would reduce the need for "one off" production, and utilize the economies of scale that modern manufacturing is capable of delivering.

Smaller standard designs would make it easier to handle disruptions, make a power grid more resilient in the face of natural or man-made disaster. It would also make it easier to scale. Having a power grid with interchangeable parts would reduce the logistical complexity of spares management.

There are already standardized transformer sizes(e. g. 40MVA). Electrical transmission is pretty standardized already IMO.

The problem is that large scale transformers are expensive to produce which means that it makes no sense to keep stockpiles for a disaster case. Usually that calculation works out, as most electricity grids, even on the low-voltage distribution level, are meshed enough to tolerate one or sometimes two large-scale transformers going offline.

You get into problems though if more of these transformers go offline - e.g. because someone shoots them with rifles or artillery - because the lead time is weeks to months. And completely forget about a nuclear EMP scenario... it would probably take a decade to recover from that.

Transformer protection equipment is quick enough to disconnect them in case of an EMP, it's usually the control electronics that are identified in vulnerability studies as the weakest link. Those have a much shorter lead time than the multi-ton copper parts and are also much easier to have a stockpile of.

They are standardized, they're just not "mass produced" due to difficulties in automation at that size.

I suspect transformer production could be ramped up if the manufacturers were convinced that it wasn't just a temporary blip.

Automation isn't worth it unless you have a high demand. A skilled machinist can create many different parts, one day he makes parts for a transformer, then next he is building parts for a ship, and a week later he is doing custom parts for the rail roads. A skilled automation engineer would still need a few months to automate each of those separate parts, while the machinist is is done in a couple days with each. Once the automation is done the automated process can produce a lot more parts a lot faster, and cheaper - but if you only need one/a few the automation isn't worth it.

Of course automation is getting better/cheaper. Things that were not wroth automating in 1960 are now so easy that you would automate it even if you only need one. Modern CAD/CAM systems can often go from drawing to automated production with the press of a button. Over time more and more things will be automated.

Skilled machinists need to be trained. Sometimes you automate production just because you can get an engineer to design the automation faster than you can get access to skilled machinists to do the job. (this assumes the automation can be set up with less skilled labor - often automation itself requires more machinist time than one of the part it is making)

Standardization isn't really the issue described here. One of the issues is the manufacturing capacity for producing standardized parts is limited. The manufacturers are not interested in investing in the costly expansion of capacity required to meet a temporary surge in demand.

Other issues are regulatory in nature. The article goes into some detail on the matter.

The issue is not that much about the ingress or about a single piece of equipment, as it is that adding new power sources (or consumers) creates topological changes in the network.

That new sources changes the power flow not only locally, but potentially all over the grid. It also changes the way TSOs react to equipment failures: typically, when an equipment like a power line opens, the power flows differently in the network. In order for other equipments not to become overloaded, operators study the network situation and have go-to solutions that involve changing the network topology, rerouting power and, in extreme situations, cutting some consumers. If power sources within the network change rapidly, that study work becomes obsolete fast, and that becomes a problem for operators and their ability to maintain a secure transport grid.

TSOs are working on it with both R&D and industrial applications, but that's not a fast process, and sometimes it challenges assumptions that used to be safe, and uncovers big question marks.

The point I failed to make is because I brought a machinists understanding to an electrical engineering conversation. I assumed a specific definition of the word "standardized" that isn't common to electrical engineering.

Here's what I should have said....

What's needed is standardized, interchangeable transformer modules. We need to be able to parallel as many of them as we need to manage a load, and if one dies, swap it out with a new module, possibly 30 years later, with NO need for adjustment.

Just as precision makes it possible to deliver parts that fit within 100 microns from countries on the opposite sides of the globe (and has for almost a century thanks to Johansson[1]), we should be able to produce a standard, easily paralleled transformer module that can simply be bolted to others to match situational requirements.

Bolting in a new module shouldn't result in the need for balancing or any other adjustment, the parts should just work together.

This would have helped immensely in Ukraine, for example... as missile damaged sections of a transformer stack could be individually replaced, instead of the whole thing.

There are enough CNC tools and measurement techniques in the world to make this possible, we just need to make it happen.

[1] https://en.wikipedia.org/wiki/Gauge_block

In the US grid capacity doubled from 1960-1970 and doubled again from 1970-1980. We can do it again :)

Did you actually read the article, or did you just want to say something to try and force fit reality to your preferred narrative?

Did you? There is nothing in the article to refute the parents really good point.

Doubling the size of the grid was not easy any of the other times it happened. It was hard then, but it got done.

Except there has been a vast increase in paperwork and potential liabilities since the 1970s.

When the problem gets "big" enough paperwork and liabilities get bypassed; it just needs the political will to force something through.

It's happened multiple times in history and it will happen again.

In fact, if the only real problems are paperwork and liabilities; the problem is basically solved; it's so solved that we're fretting over details.

I think part of the problem is that there's a segment of the American population that doesn't want the grid to expand. They get to point to the grid as a excuse to curb EV adoption. And all politicians have to do is nothing to make it happen.

If the options were expand the grid, or give up air conditioning, you can damn well bet the politic winds would hit hurricane speeds.

https://news.ycombinator.com/newsguidelines.html#:~:text=Ple....

Let's not forget that solar panels or roofs and local batteries (EV cars) *reduces* load on the grid. Hyperlocal power generation does that.

In theory and absolute metrics, yes. As you said “clean” energy is associated with “clean” consumption, meaning more EV, electric heating, etc. Those local power generations helps supporting the growing needs for electricity but don’t change much to the currently running industry and residential usages (oven, dryer…)

There isn't a growing need, at best it's a want.

tl;dr Connecting new power generation to the electricity grid is a queue-based permit system that is stupid and slow, but simple and supremely stable. There is a decent chunk of renewable power “in the queue” but speeding up the connection process risks damaging stability.

The article notes India’s electrification as a partial counter-example: relatively fast, but correspondingly relatively unstable. (One marker mentioned was “kerosene liters consumed”, as a proxy for how much lighting has been replaced by electricity, they note the amount dropped from 9 billion liters to 2 billion liters, which I take as a very rough indicator of ~80% grid reliability - not even one “9“ of stability. It’s hard to find representative data on developed-world power grid reliability but for instance there is a common estimate that the average Australian resident experiences 200 minutes of power outage per year, which corresponds to somewhere above three-and-a-half 9s and below four 9s. This suggests potentially quite an extreme cost in stability for the moderate benefit of speeding up queue times for renewable power.)

If I understand correctly there's little slack in transmission capacity to service new supply, and that new transmission infra isn't built out until new generation is queued. What wasn't clear to me is if construction of the new generation has to be completed before it can be enqueued or if e.g. completed permitting is sufficient.

Anyone significant planning on new generation will also plan the interconnect at the same time, and that includes transmission capacity. There is no point in having your new generation completed if the interconnect isn't done. In fact the bank shouldn't (and probably is checking, though not always) let you get financing for the new generation if you have not verified that there is interconnect to use that generation. Thus plans for transmission is supposed to happen before you get to permits. Though depending on timelines you may start building on before the other.

You putting solar on your roof is not significant. Your whole neighborhood putting solar up should be significant, but odds are this is done by individuals who don't realize their collective action is significant.

The average outage time per year in Germany, as far as I remember, is below 2min. The uptime of my non-USV-backed server at home supports that, but that’s only anecdotal data.

Fellow German here, I’ve experienced a total of ~2 hours, in one single outage, about 20 years ago. Not even a single brownout since.

In my experience, USVs can make the uptime worse in Germany. The hardware is likely to fail before the next blackout.

According to tennet.eu (the grid operator for the Netherlands and parts of Germany), the uptime for their high voltage grid is about 99.99963%. Obviously it will be lower for individual consumers, since there is a lower voltage grid in between them and the high voltage grid that might fail separately. Still, I think I've personally experienced only one power failure in the last 5 years or so.

The only power cut I've experienced since I've been in the UK (8 years) was about 30 seconds long. By the time the operator had sent me an SMS about it, it was already back up. This is despite the UK's rapid transition to renewables. There was a fairly major power cut by our standards in 2019, but it didn't affect me.

Personally speaking, power cuts are so rare and short that I don't think about them or plan for them.

Here in South Africa the power can be off for anything from 2h to 8h a day and given the regularity of that interruption I've since gotten an inverter (with batteries) as well as several battery powered emergency lights to power me through. I know that many residents of India do something similar, and so I'm a little skeptical of this kerosene marker.

A honest, but offtopic question (if Dang sees it fit to remove, let him do it...): why do you stay in SA at all?

What are the redeeming qualities of the country, aside from the nature, which is fabulous?

If you are white, aren't you afraid of a possible pogrom against your family?

I'm not OP but it's incredibly hard to immigrate and as older people are less mobile/flexible it may mean leaving your elderly parents/grandparents behind, as well as your native culture and language and everyone/everything/everywhere you know.

I am thinking about leaving SA - the usual factors of cost, family, and change apply.

Redeeming qualities: I do have pretty good buying power in SA, the weather is great,and it is quite a multicultural place with friendly people by and large.

To my knowledge the rate of black on black violent crime is higher than black on white crime - SA has a general problem of violent crime. Its definitely a problem but a pogrom specifically is not a chief concern for me.

Thank you for an interesting reply. I come from a fairly safe country, so I cannot really imagine how living in an unsafe country looks like.

What precautions do people in SA need to take when travelling? Is there any sort of constantly updated "off-limits area map", or is it just common knowledge?

How does security in wealthier neighbourhoods look like?

Is hiking in the wild risky? Is public transport generally to be avoided, or are there "better and worse" services?

As far as electricity blackouts go, everyone has a generator, right? Is there never a shortage of diesel for those generators? What about the pollution that comes from running generators several hours a day?

> Is there any sort of constantly updated "off-limits area map", or is it just common knowledge?

No maps - just common local knowledge. Some areas you avoid completely, some you just avoiding walking or driving through late at night.

> How does security in wealthier neighbourhoods look like?

Most new higher-end developments are fenced-off security complexes - electric fencing all around, and 24/7 security services. But there are still many "regular" suburbs without that. In those you typically have security patrols, and individual houses may or may not have electric fencing, depending on the area.

> Is hiking in the wild risky?

Depends on the area. In some areas you just avoid hiking alone. The more remote areas typically have no safety issues.

> Is public transport generally to be avoided, or are there "better and worse" services?

Most (but not all) wealthy people avoid local public transport ("taxis" and trains) - they do have safety (and reliability) issues. There are good options in some places - e.g. a high-end train in Johannesburg & Pretoria, and Uber is a decent option in most cities. Long distance busses and trains are also typically fine.

> As far as electricity blackouts go, everyone has a generator, right?

No, generators are mostly only by businesses. Too noisy and too much effort to use for most residences.

The cheap solutions involve getting a mini backup power supply for your internet router, and some battery-operated lights. That + some planning around e.g. cooking gets you through most outages.

Recently, battery backup + solar became very popular for everyone who can afford it. Prices for solar and batteries dropped substantially over the last 5-10 years, and the entire installed system costs around 1/3rd of what you'd pay in the US.

> Is there never a shortage of diesel for those generators?

We've had one shortage that I remember over the last couple of years. It's much more relevant for vehicle fuel than generators.

> What about the pollution that comes from running generators several hours a day?

Much less than that of cars and our coal power stations. Noise is the biggest issue.

"Most new higher-end developments are fenced-off security complexes"

Thank you for taking time to write such a detailed answer.

The security complex thing, I wouldn't be able to live like that. It is my natural instinct to roam cities on foot, walk through the streets and the parks, observe people and birds etc., regardless whether it is my own city or a city that I visited as a tourist. But it seems this is precisely the one thing you can't do in SA safely.

Living behind a high wall, I would feel like an expensive prisoner. But maybe I wouldn't miss the freedom to roam, if I never experienced it in my life.

Yes, that's also why I don't live in a security complex. Many people don't, but there are many people who just want that extra safety factor.

Also, the higher-end security estates are massive - you effectively have your own park (or golf course) in the estate, so that may lessen some of that. Those places are expensive, but probably still much less than a comparable house in the US or Europe.

And in cities there are places where it's safe enough to just walk around like you mentioned, but safety is something that's typically on your mind wherever you go.

But not being able to just go where I want without considering safety, and not having my kids being able to just walk around or use public transport without fearing their safety - that is a big factor in making me consider emigrating.

The only safety conversation I remember having with our kids was about how to behave when pulled over by cops - in US.

I am sure we discussed how and where to go when they went to NYC, and that usually with friends some of whom are familiar.

Traveling wise I think it is common knowledge where the more unsafe areas are. There is always some overhanging tension living in SA, particularly in the more urban areas.

Middle class and up households often pay private security companies who patrol suburbs in branded hatchbacks and respond to alarms, etc. Robberies still happen regularly, but it helps. Complexes and gated communities are also more common. Almost all windows on houses are barred and people have security gates for their doors, and many properties have high fences, alarm systems, etc. (which don't really work because of the power problems but c'est la vie).

Hiking wise: it depends on the area and how many other people are hiking I think. I think that also applies to walking around generally.

Some households and businesses have generators, but they don't provide a high quality current, are noisy, and require diesel. Other places have inverters and batteries (like me) which charge during off-peak hours and act like a UPS when the power goes off. Quite a few households are going with solar roofs with an inverter and battery. All these things are fairly pricey even for middle class families, and quite a large percentage of the population is quite poor - so I don't think most people do have a good solution - but internet services are fully battery backed at least and many people have gas power for cooking (which helps).

Quite a substantial part of the SA grid is backed by diesel - it is worth bearing in mind that we are geopolitically quite "neutral" and still do business with Russia (who are a trade partner under BRICS - south africa is the S). Air pollution here is higher than in most US cities I think but not as bad as some other developing nations.

I've spent some time in Panama in one of those gated communities and it always felt like an 'inside-out' prison to me.

Even in developed countries I've noticed a marked difference (albeit nowhere near reported SA or India levels) when moving from Israel to Austria and then to Germany.

When I was growing up in Israel (1983-2005) power outages were common but not frequent (once every few months, my parents had a permanent supply emergency lighting, flashlights and candles at home for when the power would go out, usually for just an hour or 2).

Ever since moving to Austria (2005) and then Germany (2013) I think I have only once or twice experienced power outages and they only lasted a few minutes. So probably >20x less common (once a decade instead of a couple times per year) while also getting fixed a lot quicker. I don't know how much the infrastructure has improved in Israel since (I would guess it's better now but not as good as in Austria and Germany) but even back then it was still just an occasional nuisance unlike in the developing world where it happens enough to really impact people lives and productivity.

I think reliability should be divided up in different ways.

Their is "how well the moving bits are moving" which is different than "lines getting knocked down by acts of nature".

So, for example, I live in Seattle. Our power generation is quite reliable, but trees falling down during wind storms and knocking out power lines is also reliable, gravity being what it is.

I don't consider the Seattle grid to be unreliable, though I do acknowledge that digging up the entire city and burying power lines would, at great expense, prevent nearly all power outages.

But that sort of unreliability feels different than brown outs.

My utility concluded that between underground animals chewing wires, idiots with backhoes not locating before digging, and the extra time effort to repair burred lines: overall overhead wires are more reliable than underground. Note that they put a lot of effort into tree trimming near their overhead wires, which made trees falling on them was a rare problem.

If you want to really analyze this you have to look at outage distributions (both size and duration). High voltage line disconnecting and leaving half a city in the dark for an hour is weighted-outage-time equal to one town sitting without heating for a week. However, it would be quite hard to call such disconnections equal.

Centralized grid instability is not going to stop new power generation from coming online. Decentralized power generation is what's already happening in countries with inadequate grids. People install batteries, solar, wind mills, generators etc. Anything they can do to mitigate against loss of power or unacceptably high prices for power. The more unstable the grid is, the more people do this. Just look at most of the developing world. There are lots of countries that never had much of a functioning grid where economic growth is creating demand that cannot be met by their electricity grid. In most of these places, people just go ahead and fix things themselves. And that increasingly means solar panels and batteries rather than expensive to operate generators.

In the UK, and elsewhere, there are a lot of home owners that have solar on their roofs. The reason is not power outages but electricity prices. And with recent price spikes, more people are considering doing that all over the world. The whole assumption that most new generation has to come from the grid is increasingly less true. Both companies and consumers invest in cheaper private power generation. Especially companies have a big incentive to reduce their cost. The more power they need, the bigger the potential savings. And of course increased demand with a limited supply creates price spikes as well. We saw that in the last few years. This just speeds up the decentralization.

Australia is a good example. About a third of the houses have solar panels already and the building codes are being updated to require solar panels for new construction and renovation projects. Millions of house holds generating tens of kwh every day is adding up to a lot of power. And a lot of it goes straight into the grid as well. Which adds to the instability. Lots more will be coming online in the next few years.

Rural Canada is also a good example. The grid is horrendously bad and it is pretty easy to outperform the grid reliability wise with a local solar/wind/battery setup. You might even go off-grid completely assuming a large enough battery and maybe a backup generator just in case.

Interestingly enough, Tesla just released their "Master Plan 3" which admits the need for hydrogen for seasonal energy storage and synthetic fuels. It doesn't admit the need for nuclear though. So we can see it as a first step to coming to reality, but it is not there yet.

I bring this up because I am constantly annoyed by those who really haven't thought through their green energy dreams, and forget that compromises need to exist. These compromises actually have no serious penalties whatsoever BTW, it just requires a more complexity thinking process.

And in this case, it means that there will be a lot of times where the solution is to build a gas turbine running on hydrogen, ammonia, synthetic fuels, etc., or build a nuclear reactor. Just piling up more and more renewables onto the grid isn't a solution. We are finally seeing some people come to reality on this, and I expect even further shifts to more realistic thinking in the future.

> These compromises actually have no serious penalties whatsoever BTW, it just requires a more complexity thinking process.

Not entirely true, storage does cost money:

https://www.energy.gov/sites/default/files/2019/07/f65/Stora...

That said, as the tables in that report indicate its quite a complex topic.

(By my back of the envelope maths switching UK to fully wind and storage would triple electricity prices)

Relative to the alternatives, it does not cost more money. You just need to accept the existence of hydrogen-based energy storage and perhaps synthetic fuels too. Cost will be the same or less as other energy storage ideas. And if you accept nuclear, you don’t even need much energy storage at all.

Also, a big chunk of the “energy storage” solution is utilizing biofuels. Like in Germany where wood burning power plants are seen as green (hint: they’re not). Getting rid of this will be a huge boon. The compromise in question will have significant environmental benefits. It’s unfortunate that so many are blind to their own bad ideas and have not noticed the problems of biofuels.

I encourage you to do the maths on this a write it up in a blog post, it'll be enlightening :)

Every back of the envelope calculation of this makes the same basic mistake of acting like every watt is sacred and that spending billions on barely used batteries to save 1 Watt every 5 years makes sense.

It's just silliness disguised with maths. Which is excusable the first 10 or so times its done, but is getting old now.

What's really getting old is claiming renewables are cheap and easy to build while not calculating all the additional requirements that are needed to have a stable power supply such as overbuilding, storage etc.

That is a fair point, but what are you actually advocating for?

Tainting public perception of renewables with cost-related FUD more strongly?

IMO the need for storage and better grid connectivity gets brought up in every discussion about electromobility and renewable power already anyway, so everyone seems well enough aware of it.

I'd also like to point out that glossing over this is SUPER comparable with assuming >80% capacity factors for nuclear plants which the anti-renewable crowd always eagerly does.

> Tainting public perception of renewables with cost-related FUD more strongly?

Or maybe bring actual discussions into public perception, and not just the unquestioned undebatable "renewable everything will solve everything by magic"?

> so everyone seems well enough aware of it.

Of course very few are aware of it.

All the discussion is doe-eyed "we just need to replace everything with renewables". No one talks about the need to overbuild, and how much. Storage is assumed a solved issue even though it's not anywhere near the required scale.

They might be aware of it, but actual data is always useful.

Electrification, not renewables, is the problem to solve that requires major thinking outside the box. Your faster horses won't get us there.

The compromise is going beyond electrification. You need large scale energy storage or baseload power, plus the needs of industry (aka the hard decarbonization sector). This cannot be solved purely by electrification. People who have thought hard about the problem have realized this. Unfortunately, quite a few people are trapping in an obsession with ideological purity and haven't figured it out. Breaking this obsession has basically no negative consequences, as it simply gives you a real path to zero emissions.

Grid stability is essentially a different thing than local power availability. You lose your power if a tree falls on your power line, or if your local transformer fails. Sad for you, but completely unrelated to grid stability.

Grid instability happens when load and supply are no longer matched. That's when serious things have to happen - load shedding and rolling blackouts. In the developed world we put a lot of money into avoiding this - pumped hydro, battery storage, fast response gas turbines, frequency control programs running with big industrial users, and (opt-in) load shedding of big industrial users. All so that power is available to an end user who needs it.

As someone living under PG&E, I'd love improvements to both. In the late summer we often lose power due to under-supply, in fall due to concern for fire, and in winter due to wind. In one recent month I was out of power for eight days in three tranches.

PG&E is an excellent example of grid failure through regulatory failure. California has made a mess of electrical regulation over the decades, and today you have to suffer through the result.

I split my blame with roughly 2/3 going to CPUC and the rest going to PG&E.

Yeah, like I said it’s really hard to find representative data on grid stability. Local power outages will definitely overestimate grid instability by a lot, but it’s not entirely unrelated, it should capture ~most of grid stability issues (it is a gross over-estimate because maybe like 1%-5% of the total local power outage experienced is due to grid instability). So maybe the electrical grid is more like 5 or 6 nines.

> "speeding up the connection process risks damaging stability"

The UK has transitioned to renewables perhaps faster than any other major economy in the past decade[1], yet has a very reliable and resilient grid[2].

[1] As recently as 2012, around 45% of the UK's electricity was generated from coal. Now it's almost zero, with the last coal power plants to close by 2025. In the same period, renewables have grown from <5% to over 40% of grid supply.

[2] "Five 9s" transmission system reliability of supply nationally in 2021-22: https://www.nationalgrideso.com/industry-information/industr...

Well. I suppose this stability is mostly thanks to (A) natural gas and (B) biomass, aka wood pellets being burnt in former coal power plants. Those two basically replaced the coal that was phased out.

See this plot:

https://ourworldindata.org/grapher/electricity-production-by...

Agree, the speed of transition has been quite fast. As you can see in the following plot the UK started out at a rather low level of renewables. At this relatively low level the transition is probably easier than at higher levels.

https://ourworldindata.org/grapher/share-of-electricity-prod...

I'd guess the real challenge with grid stability will become more visible when grids all over the world reach some 2/3..3/4 of renewables.

Good management must also be counted as a factor, I think. Whatever that entails.

I don't think that chart really makes your case since it's measuring total production and on a total production basis it simply isn't true that gas and biomass replaced coal.

If you pull that chart to its full timeline, over 200TWh of coal production existed, and it's essentially all gone now, but there isn't 200TWh of gas and biomass, in total they're maybe 140TWh. There is 60 TWh of wind power production.

Now, in terms of capacity for instantaneous power, the Combined Cycle Gas Turbine plants probably do add up to similar capacity to older coal plants which were phased out, at least very close, but it's pretty rare for all that generation to actually be needed - this chart doesn't really show that.

As to "transition" the countries which look "slower" on this chart are mostly using a lot of hydro, which is very different from wind or solar, as well as being readily available from the mid-20th century a point where solar and wind power were not really established options. So we're not talking about China having a bunch of wind turbines, then kicking back for a decade and only adding more recently, but instead their enormous country has hydro power, and now it also is adding wind turbines. The UK is a small island (and some other even smaller islands which are useful for wind but don't produce hydro power) and doesn't have vast majestic water bodies like Lake Mead, so hydro power has always been a tiny niche here, and that's all you're really looking at.

> If you pull that chart to its full timeline, over 200TWh of coal production existed, and it's essentially all gone now, but there isn't 200TWh of gas and biomass, in total they're maybe 140TWh. There is 60 TWh of wind power production.

Indeed, if you view the whole chart, the development during the 90s directly supports my claim I'd say.

https://ourworldindata.org/grapher/electricity-production-by...

But still, I guess I phrased my actual point rather poorly. True, wind and solar produced almost 1/3 of the UK's total electricity over the last 12 months, but what kept the grid stable was mostly natural gas.

In this role natural gas replaced coal.

> Now, in terms of capacity for instantaneous power, the Combined Cycle Gas Turbine plants probably do add up to similar capacity to older coal plants which were phased out, at least very close, but it's pretty rare for all that generation to actually be needed - this chart doesn't really show that.

Yes, pretty rare. But this is exactly what grid stability is about. Making sure the lights stay on even though it's cold and there's no wind. As for example end of November 2022 till middle of December:

https://www.electricinsights.co.uk/#/dashboard?period=1-mont...

Hydroelectric power was pretty widely available by the late 19th century.

Another interesting example is Southern Australia. It also went from coal to renewables, and also with gas as "backup". Its just fascinating how far they have come, from less than 1% renewables in 2007 to more than 60%, all without any hydro which makes reneables "easy".

https://www.solarquotes.com.au/blog/sa-renewables-milestone/

> "I suppose this stability is mostly thanks to (A) natural gas and (B) biomass, aka wood pellets being burnt in former coal power plants. Those two basically replaced the coal that was phased out."

Actually, most of the coal has been replaced by wind and solar. Biomass is not really that significant: it's only around 5% of the UK's grid supply, compared to over 30% for wind and solar. New biomass plants are no longer considered renewable in the UK so it's unlikely to increase in the future.

Not sure what data your chart is based on, but it's wrong, at least for biomass. See here for some up-to-date data (you can use the controls at the top to see different periods and historic data): https://www.electricinsights.co.uk/#/dashboard

Natural gas is still the largest single contributor to the UK's grid supply, but it's in decline, and will be overtaken by renewables in the coming years. Wind capacity will triple in the UK by 2030!

Imports have also increased in recent years as new HVDC interconnects have been brought online.

> Actually, most of the coal has been replaced by wind and solar. Biomass is not really that significant: it's only around 5% of the UK's grid supply, compared to over 30% for wind and solar.

True. By electricity produced natural gas and biomass have not replaced all TWh that formerly came from coal. (Even though you can see how natural gas directly replaced coal during the 90s here: https://ourworldindata.org/grapher/electricity-production-by... )

The point I intended to make is: there's two kinds of electricity production capacity. There's Flexible/dispatchable capacity such as gas, coal and others. And there's intermittent capacity such as solar and wind.

Without electrical storage (almost non existent), solar and wind cannot replace dispatchable capacity in the context of grid stability.

The increase in electricity produced from natural gas and biomass is what directly contributed to the grid remaining stable, _despite_ the increased production from wind and solar.

> Natural gas is still the largest single contributor to the UK's grid supply, but it's in decline, and will be overtaken by renewables in the coming years. Wind capacity will triple in the UK by 2030!

Yes. But it won't be the wind, that'll be keeping the grid stable, that's for sure.

Natural gas did partly replace coal during the 1990s, but renewables (mostly wind) is replacing natural gas today.

An increase in production from natural gas isn't needed to keep the grid stable. It just needs to be available for when supply falls short due to weather conditions.

The interconnections are also a big part of the solution: excess wind energy can be exported when it's plentiful, and imported (along with hydro, nuclear) from other countries when UK weather is unfavourable. It's always windy somewhere!

> "But it won't be the wind, that'll be keeping the grid stable, that's for sure."

No, but it will greatly reduce emissions, and improve energy self-sufficiency, security of supply and stability of prices when external events (like the Russia-Ukraine war) disrupt the market.

> Well. I suppose this stability is mostly thanks to (A) natural gas and (B) biomass, aka wood pellets being burnt in former coal power plants. Those two basically replaced the coal that was phased out.

More like a little under half due to those. From the plot at ourworldindata, expanded to start at 2012 since that is when the comment above was talking about, here are the changes in TWh contribution:

  -137 Coal
   -23 Nuclear
     0 Other renewables
     1 Hydro
     3 Oil
    11 Solar
    24 Gas
    24 Bioenergy
    45 Wind

This data does not seem consistent with other sources. There are no active oil-fired power plants in the UK, for example, and there hasn't been for a long time. (There are some that were built as oil-fired plants but were later converted to natural gas CCGT)

ourworldindata also seems to be overstating bioenergy by a huge margin. According to reliable sources[1], biomass supplied 20.83 TWh in 2021, not 39.11 TWh. I'm not sure what difference in methodology could account for such a big variance?

[1] https://www.electricinsights.co.uk/#/dashboard?period=1-year...

> One marker mentioned was “kerosene liters consumed”, as a proxy for how much lighting has been replaced by electricity, they note the amount dropped from 9 billion liters to 2 billion liters, which I take as a very rough indicator of ~80% grid reliability

I don't think those are related at all. Someone without electrical infrastructure running to their village is going to burn fuel regardless of how reliable the grid would be.

Kerosene was subsidised heavily as it was used for cooking stoves widely.

At some point kerosene subsidy was removed and people shifted to battery based backup systems/ lights etc.

Electricity is unstable in large parts is my experience.

The kerosene they are talking about was for burning in backup generators, not lamps.

Grid power dwarfed that produced by local generators even when it was 9 billion litres. 80% stability isn't the right conclusion from this.

here in CA, the CPUC is trying their best to prevent people from adding Solar to their roofs. They've even got rid of Net metering almost entirely with their new NEM 3 rules.

Net Metering was a subsidy to encourage solar growth at a time when it wasn't nearly as prevalent as it is now. It's not the "natural" way of metering solar-equipped homes. (Assuming there is any "natural" way at all).

Now that solar is cheap enough to support itself without subsidy, the math is changing at the grid connection.

Everyone is producing gobs of power in the middle of the day, and just expecting the grid to take anything they can't use onsite. Then in the early evening, it all reverses quickly and the demands on the grid quickly flip.

In my mind, a connection to the grid should pay per kWh for two things: generation and transportation. When I pull power from the grid, I pay for both of those things. I pay to have those watts generated, and I pay to have them sent to me.

When I push power onto the grid, I am paid for the generation. I do not get paid to send it to wherever it goes, that's someone else paying that part.

Nice survey of challenges. For follow ups, deep dives on the manufacturing of grid hardware and software would be cool.

For a (mostly) US perspective, check out David Roberts' Volts, "a newsletter about clean energy and politics". https://www.volts.wft Links below to episodes specifically about grids.

But since our grid is the elephant in the room, it's touched on in most episodes.

My noob TLDR for USA is:

Overlapping jurisdictions are a huge roadblock. To build new capacity, you likely need permits and buy off from every state, county, property owner, and special interest touched.

There's no federal plan to reform our currently siloed systems. Build Back Better addressed this. But because the Inflation Reduction Act was passed thru "reconciliation", it doesn't contain those "third leg" of necessary reforms. Huge disappointment.

Predictably, progress is being further stymied by a huge reactionary anti-electrification noise machine. All the usual suspects are dumping money into astroturf groups and propaganda to oppose anything and everything, from windmills to induction stoves.

NIMBYs have weaponized environmental regulations, created in response to past abuses, to thwart progress.

Just like how "the internet treats censorship as damage and routes around it", reformers and innovators are finding alternatives. Stuff like: colocating generation with consumers (industrial heat), embracing geothermal, and beefing up existing grids with storage.

Transmission week: why we need more big power lines https://www.volts.wtf/p/transmission-week-why-we-need-more#d...

Transmission week: how to start building more big power lines https://www.volts.wtf/p/transmission-week-how-to-start-build...

Transmission fortnight: burying power lines next to rail & roads to make a national transmission grid https://www.volts.wtf/p/transmission-fortnight-burying-power...

Transmission month: how to make the existing grid work better https://www.volts.wtf/p/transmission-month-how-to-make-the#d...

Transmission month: two more ideas to quickly boost the transmission grid https://www.volts.wtf/p/transmission-month-two-more-ideas#de...

The challenges of building transmission in the US, and how to overcome them, with Liza Reed https://www.volts.wtf/p/volts-podcast-the-challenges-of-buil...

What's up with Manchin's plan to reform energy permitting? https://www.volts.wtf/p/whats-up-with-manchins-plan-to-refor...

[Bay Area] Peninsula Clean Energy attempts to achieve 24/7 clean energy https://www.volts.wtf/p/an-energy-provider-attempts-to-achie...

Utilities are lobbying against the public interest. Here's how to stop it. https://www.volts.wtf/p/utilities-are-lobbying-against-the#d...

People are so wedded to fossil fuel religion, they'd rather pretend their country is entirely incapable of the most basic tasks.

China isn't playing by the pretend limitations you put on yourself and it looks like you've already admitted they deserve to be the new hegemon. Yay for totalitarian communism I guess.

The stat that 80% of the UK queue might be effectively domain squatters with no actual project waiting to flip to real developers is shocking. Pure rent seeking middlemen.

Though at the same time, it means all the other stats are BS. Like saying that a concert is sold out and there's no way to get a ticket until the next time they visit in 5 years time because 80% have been sold to scalper's bots. It just doesn't logically add up. The tickets are available, you just need to pay a markup to a scalper who performs no real service to society.

Lol yeah, I read the title thinking “yeah obviously, nothing at population wide scale is a simple task”.

It’s a good thing we are pretty good at performing all kinds of not simple tasks and have been for a very long time.

> have been for a very long time

Aren’t we good at this especially since fossils fuels ?

"Yay for totalitarian communism I guess"

Unless the government in its infinite wisdom decides to exterminate all sparrows or so and causes a countrywide famine, I guess...

Some of the limitations are pretend, but some are very real. European countries in general want their grid to be stable and reliable, and the surges and dips of power that come from "too much sun" or "too much wind", or lack of either, are going against it.

On the other hand, India, mentioned in the article, has so many blackouts already that it can cope with some extra instability. Of course, the unreliability of the grid contributes to the reluctance of foreign investors to build industrial plants in India, which is a major downside, but hey, they can alter their grid faster.

That entire framing of "renewables == unstable grid" is the false premise this entire article rests on. It's just out-of-date, proven to be wrong by decades of experience FUD.

"Too much" solar and wind are not, and have never been, a problem for a grid.

Almost. Along the German border there are a couple of HV interconnects and those definitely caused trouble on the other side on high wind days.

https://www.google.com/search?q=germany+czech+grid+windfarm+...

Most of this has been dealt with in the meantime, but such problems really do occur, but they're utterly blown out of proportion by the anti-renewables lobby.

> He said Germany could solve the problem by allowing its grid operators to turn off renewable sources at times of excess production, which is not currently possible under German law.

This is not a technical problem, but a legal one. And even the legal issue has at least 3 technical solutions, they're just negotiating over who should bear the cost of working around the legal situation.

The issue isn't that the grid can't handle large amount of renewables, it's that renewables aren't consistent.

Large thermal plants provide inertia that renewables simply can't provide. A lot of the work around incorporating renewables into grids is around the provision of synthetic inertia and how to manage sudden drops in supply.

Battery storage is one solution. Better demand response solutions backed by energy markets that incentivise the participation of large numbers of energy users is another.

Grids that have a high level of interconnection with other grids also reduces the impact of inconsistent supply, but not every grid has enough inter-connections for that to be a viable full solution. (See ERCOT in Texas, AEMO in Australia, Eirgrid in Ireland for examples)

Here in Ireland, we frequently run the grid up to 80% renewables, but the grid operator only does that because they have the mechanisms in places to handle sudden drops in supply.

well, not so fast. so far China managed to build a lot of things in a lot of places, but increasingly they are facing a problem of population-generation mismatch, so they would need to build a lot of cross-country transit capacity, which is not cheap, not simple.

and that's also one of the reasons they still have so many coal fired plants close to population centers.

and we'll see how long Xi's luck lasts (which in practice means how long the benefits of the reforms will last, before the inevitable corruption of dictatorships eats it up)

also, tiny nitpicking, so far it seems China is just very authoritarian, but not totalitarian like North Korea.

China is a world leader in HVDC and UHVDC transmission and has been for over a decade:

https://www.power-technology.com/features/chinas-mega-transm...

> The Changji-Guquan +/-1,100kV project is a major technical step-up, making it the world’s highest DC voltage,” said Liu ZeHong, executive vice president of SGCC. The project is the first of its kind, but could be followed by many more crisscrossing Asia and the world. Already SGCC has plans for the construction of more ultra-high-voltage DC (UHVDC) lines both domestically and as part of its Belt and Road Initiative, a project that is aiming to increase Chinese-led infrastructure investment in more than 80 countries.

> SGCC president Liu Zhenya has dubbed UHVDC an “intercontinental ballistic missile”. It is a key part of his Global Energy Interconnection initiative,

Its amazing what you can achieve when the main conversation isn't "are all the scientists hoaxing us because they irrationally hate our precious fossil fuels?".

It's an outright trivial task.

This can tip the scale in favor of efuels (as opposed to battery-electric). Just run the electrolysis straight from a windfarm and produce at the rate the windfarm is currently outputting.

The very low efficiency of e-fuels probably more than outweighs the cost of upgrading the grid. They will be important for sustainable aviation, but anywhere else they don't appear to be economically feasible.

This is what’s currently happening. It’s windy in some remote places. Connecting those places to the grid is horribly expensive but building a chemical plant there isn’t. That’s the locations where efuels are being made today. Like the Porsche project in Chile.

https://newsroom.porsche.com/en/2022/company/porsche-highly-...

An idea that sounds good, but has the problem that efuel production capital expenditure is fixed and high. So once you've built it, you want to run it 24/7 rather than intermittently.

I'm very much open to suggestions for "low capex ways to make productive use of 'spare' electricity, preferably reversibly even if efficiency is low".

Reply on news.ycombinator.com