As one of the comment threads on the Youtube video mentions, this setup of having tiered power generation comes up naturally in the video game Factorio. You have mining drils feeding coal to boilers, and the boilers feeding steam to steam engines. If the power runs out, the system can't start back up because the drills need power. So you end up automating the setup so that if the steam reserves drop too low, the rest of the factory gets disconnected from the steam engines until it builds back up.
Furthermore, some mods make it so that even pumping the water to the boilers requires power (*), and add a separate "burner generator" that directly burns coal to produce power. So you end up setting the automation so that tier one power generation uses the burner generators to build up water reserves, then tier two power generation turns off the burner generators and uses the boilers to build up steam reserves, and then finally switches on the rest of the factory once steam reserves have been built up.
(*) In the base game the water pumps magically work without needing electric power, because the devs don't want to complicate the experience of setting up power. So it's an amusing situation where the Factorio character starts with the innate knowledge of how to build magic water pumps and steam engines out of iron plates and gears, but needs to "research" more trivial things like dumping rocks into a lake to landfill it and create a buildable surface.
I mean in theory that research involves understanding how best to layer the rocks to prevent erosion, shifting or flooding of the foundation of a new section of the factory.
We were taught the basic physics of the Newcomen engine in school as part of history on the industrial revolution, but nothing about geo-engineering for land reclamation.
AAI Industries? [1] There's a separate building (Burner Turbine Generator) to generate electricity from combustion without needing water - though less efficiently.
Like I said, there's a coal burner generator that burns coal directly into electricity. Its efficiency (energy produced per energy consumed) is less than the water boiler + steam engine, so you have an incentive to switch to steam once you've researched it. But you need to keep the burner generators around for re-bootstrapping.
Another step you forgot to mention is that, in both the base game and the modded (AAI Industry) game, there's a much less efficient kind of mining drill (burner mining drill) which is directly powered with coal (or other solid fuels), instead of electricity; when mining coal with it, you usually loop back its output to its input (through a burner inserter, another device powered directly with coal or other solid fuels instead of electricity), so the only thing you need to start up the whole system is a single piece of wood as the fuel for the farthest burner mining drill.
There's a few different mods which do this and it depends on which one how they solve it. Some start you off with power in the first place in the form of wind turbines, and some give you an inefficient way to produce power from fuel which doesn't require water.
Perhaps they have very long tendrils extending into the ground that can allow them to use geothermal energy! I'd usually say built-in solar and batteries, but they work even if you place them at night.
There's a game called Satisfactory that I've played that is kind of similar. When I played, there was no way to automated those breakers so I'd cut the lines to my power grid when I had an outage. I'd then have to fire up my biofuel generators to kick start everything including the coal miners. lol. Loved that game.
You did need to power the water pumps and it sucked depending on the distance.
I built "power hubs" that had 8 coal generators with an included set of backup batteries which I connected a modded wind generator to (literally free constant minuscule energy) so everything could be self restarting with the flip of like one switch to turn off power distribution.
Factorio liquid handling sucks. And it's just water pumps that don't need power--because your first power generation technology is coal-fired burners and they need water to convert to steam to run the steam turbines. They gave us magic pumps rather than provide some means of making power without water.
Likewise, magic belts because belts come far earlier than power.
My initial setup is to use burner inserters (which don't need an electrical connection to move) to feed the boilers, combined with priority splitters so that the ore smelters only get whatever coal the power doesn't get. Combined with overprovisioning of steam power, it means that the factory will start shutting down before the power grid gets overheated to the point that it can't keep itself up.
Later, I move to nuclear power, where fuel use is such that even glacial low-power inserters can keep the facilities fed through a power spiral. Although my massive 12-reactor facility did rely on pumps to keep the steam moving fast enough through my flowmeter to regulate power... but when you have a 12 reactor powerplant, power death spirals take long enough to occur that you will have ample time to fix them before things start needing black start.
I haven't played it that much, I've normally worked from an overprovisioning approach. Extra power plants are a trivial cost compared to your research. Even vast solar fields aren't all that expensive.
I do agree that if you try to go for a most optimum setup you do risk getting in a black start situation.
My current game I'm trying a totally different idea--I'm trying to play it with no defenses at all, rely purely on killing biter bases before they get upset.
This article (or rather transcript) got me thinking that building and maintaining a power grid with all connected power stations, lines etc. while power consumers keep popping up automatically would make a good RTS game a la Mini Metro (ok, maybe make it a little more realistic, but not as detailed as Factorio)...
The first thing that comes to mind is ENTSO-E's Power Flow Tool. It had a number of simulations of Europe's power grid with typical demand scenario's, prompting you to take care of grid balancing when a power line suddenly failed during peak load.
This demonstration tool is not connected to our real grids: therefore you can feel free to go to the limits and test out all constellations that interest you.
South Australia had to test one for real in the big storms a few years back. "It's harder than it looks" time. Two contracted gas turbine black start sources didn't work and had pretty major contractual penalty issues.
I don't see why you can't test it: take one of these contracted sources, disconnect it from the grid, and verify it can start on its own. What am I missing?
That's testing one source; and you CAN do that but you may find that there are things you didn't realize because the grid was actually UP.
For example, the entire station may be disconnected from the grid, but somewhere in the system is a Siemens device that for whatever ungodly reason needs to access the internet, which works when the grid as a whole is operational but fails when it can't get cell service.
And even if those stations work correctly, you still haven't verified that the actual plan is operational because you haven't actually done the black start; perhaps the grid divisions are incorrect and the first load you try to apply will brown out the start source.
I agree there are other things that can go wrong, but the "two contracted gas turbine black start sources didn't work" problem does still seem like something that probably could have been avoided with regular testing?
Yeah, we certainly should be doing as much as we can, but it seems time and time again we keep finding things to fix only when it's actually forced to test.
It's the same reason that datacenters don't just run the generators every month, but actually switch over to generator power and disconnect from the grid to verify it works (and if it doesn't, hopefully the batteries last long enough to get back to the grid!).
And then during an actual outage you find out that generator conks out 15 minutes after startup because the fuel filters are clogged. Between tests enough fuel would have leaked through the clogged filters to create a 15 minute revisor but during an actual outage the filters can’t keep up.
But that's just partially true. Sure, you can't test the whole system at once. But testing each station separately is totally doable. Seems like there was no oversight to check whether the claims made by the station were factual and tested.
Probably not from one of the SS-18 variants that had a single large 20MT or 25MT warhead - don't think the Russians have any of those still deployed though.
"The power station was also promoted as a tourist attraction, with visitors able to take a minibus trip from "Electric Mountain" - the name of its nearby visitor centre - to see the workings inside the power station;[18][17] 132,000 people visited the attraction in 2015.[19] However, the centre is now closed with no prospect of reopening."
Anything like that will be widely known "in the industry" even if it's not openly described on the internet like https://openinframap.org/
And if it is known in the industry it is absolutely guaranteed to be known by all countries, friend and foe alike; and real "foes" will have even more knowledge.
What is kept secret is what tooling the various militaries may have to assist in a black start, and what would trigger them actually attempting assistance.
It's one of those almost paradoxical things - the maps by the people who use a system can be very incomplete - because the system is right there! You don't need a detailed map of where the rail line between two points is because you can just follow it.
Another example is blueprints - they seem very detailed and accurate until you realize that quite a bit is "left out" and the tradesmen are assumed able to figure it out - it will show fixture locations for example, but rarely show exactly how the wires are run to them.
But if you're creating a 'map' of a house you bought, wire locations is extremely useful because they're all behind the walls.
You can restart most UK power station using a diesel-electric locomotive engine (or two). Back when I was a train engineer we handled contracts to supply them sometimes.
I'm now wondering what the state of the art is in grid-scale solar inverters and whether the firmware would allow you to black start without having to ring up Siemens.
Theoretically solar panels in a good place to do it: can output power without needing input power, can have several megawatts on a site, can match phase very easily, can potentially generate phase.
Admittedly you have to wait until the sun comes up, but in a true black start situation things are going to move slowly.
My thoughts are that power plants don't stop completely in blackouts, they disconnect, power down, and idle. The reason they need to disconnect (in chain reactions) is that the load is too high for them to maintain frequency/keep in phase.
Starting back up is complex because loads need to be added to the network in some order they can be powered.
The generators have safety systems in place when the grid tied frequency deviates too much from 50 or 60HZ - they then "trip" and disconnect themselves from the
grid to prevent damage.
Powering them down to a complete stop is a rather long process taking hours.
One silly technician left a spanner in one turbine after maintenance, and it took about 2 years for a replacement part to be installed.
Black start is just that: powering up the whole grid from blackout. It requires stuff you might not think of like the gas turbine stations need a big spark to light the flame so a diesel generator isn't enough if the station breaker is open, you need a hv DC transformer.
Not all power plants can self-feed the power they need to keep generators working. Often this is a diesel generator on the plant because the only other source of power is 20MW generators that simply can't be tuned down enough to handle the few kilowatts it takes to keep the generators turning.
Almost all plants will trip from the grid if it gets bad and remain idling for some time. But if that happens, the grid is still dark and the black start generators are the best equipped to bootstrap the grid. You can't start with the big generators too, so you can initially only bring in small plants until the grid capacity is big enough to not trip out when the big generators come online in a wrong way.
There is a good chance that by the time the grid is stable, atleast some power plants will have had to shutdown entirely and spin back up from a black start with only external power.
Nukes will shut down in blackout conditions because of the two independent power feed safety requirement. You *really* do not want your nuke plant to run out of power (that *will* Fukushima) so when the number of available sources drops too low they shut down so as to reduce the damage should they actually go dark.
In South Africa we are experiencing major shortfalls in our power generation (coal fleet is aging and breaking down more - poor maintenance and two new mega power stations has been mired with corruption during building)
We have a few cycle gas turbines (diesel) and two major hydro schemes which maintains a critical reserve at all times of water, so they just need to open the sluices to jumpstart the system.
Also, our grid is tied to our neighbors so they can help out if need be like France does to the UK during shortfalls.
The fiber box at home is 12V so it is easy to power yourself - the local POP has battery backup since it is installed at the same location as the cellphone masts for our neighborhood.
The core routers in probably in a datacenter with battery/diesel backup.
We had outages in the beginning because they use DHCP for the fiber box to get an IP address so you can imagine a few thousand ONT terminals all coming online nearly at the same time and hitting one DHCP server
I was just thinking about how decentralization is important for building robust systems, but it’s not simply decentralization, also the ability for the nodes to operate isolated from each other.
The grid is “decentralized” but still brittle and at risk of cascading failures. A future where the majority of power is provided by onsite solar + storage is much more robust.
Similarly not robust are things like tech stacks with complex “cold boot” plans or unexpected cascading failure modes, etc.
I don’t know what the right terminology for this subset of decentralization is.
Many solar power plants come with 3-4h of battery simply to time shift output to peak demand. Because solar is DC you effectively skip inefficiency from DC>AC (Solar to grid) and DC>AC (Grid to battery) conversion. On top of this you can use the same DC>AC equipment for both the batteries and solar plus being able to convert DC>AC to charge batteries from the grid becomes optional.
Parts of Australia have this. They have so much rooftop solar (and reliable access to plenty of sun) and battery there that effectively centralized power is less and less relevant there and quite a few parts of especially southern Australia effectively run on solar and battery for most of the time. It has indeed made the grid there more reliable; not less. Decentralization is mainly bad for centralized power companies as it erodes their monopoly and forces them to deal with the notion that their power is too expensive most of the time.
Intermittent scarcity and over supply of energy (both are a thing) also leads to different consumption patterns. People valuing cheap energy when it is scarce invest in batteries that they charge when there's a surplus. Others buy at a premium or simply shift when they consume their energy. Demand shaping is also something that can be incentivized financially.
Of course that works great in Australia but other parts of the world have other options like wind, geothermal, hydro, nuclear, etc. Which is why places like Norway and parts of Canada are also mostly running on clean energy at this point.
The key is actually not isolating but interconnecting. Norway is part of a connected grid that spans most of Europe with a market where you can sell energy and buy energy. Norway is an energy exporter at this point and there are plenty of countries importing their power to supplement their own generation when it falls short. That's why countries like Germany are now predominantly wind and solar powered and yet doesn't have blackouts when there's no wind. Like today (gloomy winter day with barely any wind here in Berlin). But there's still plenty of wind elsewhere. Germany imports almost as much power as that it exports each year.
Solar and batteries have some challenges about maintaining frequency and syncrhonization. When using spinning magnets (wind/steam turbines) the frequency is related to power draw. As demand spikes, the generators get slowed down. Effectively the rotational inertia works a bit like a buffer.
Balancing the grid is hence a matter of 'maintaining frequency'. Which makes the frequency very stable.
Solar and batteries are inherently DC. They have converters that take a reference AC signal, and output their energy synchronized to that reference. Incidentally, this makes it difficult to run your house off solar during a black-out, because you lack the reference. Even in smarter converters that don't need a reference to match the frequency, they still need a reference to synchronize the phase.
When we move to more batteries and solar for generation, that will put more strain on the spinning-based generation for short-term fluctuations. Less spinning mass means less buffer. At some point, any flaws in behavior of DC converters might start dominating. If they happen to resonate with each other, we might see some weird things.
It is categorically untrue that "it is hard to run your house off solar". Thousands of sailboats, and country houses around the world do just that.
It is true that if you insist on making the cutover from inverter derived AC to grid AC without an interruption, you will need some subtle circuitry that can automatically "glide" the inverter into synchrony with the grid before cutting over.
I don't know if Tesla, Bluetti, and EcoFlow devices do this, or just make a step-change in the phase and frequency of the output. That would be an interesting investigation, maybe Matthias Wander can do it?
I feel like the bigger issue would be lack of feedback, since presumably an inverter can maintain perfect frequency under any load right up until it can supply Lo more power; I suppose voltage measurement might become more relevant than frequency in such cases (or perhaps inverters are programmed to lag a bit under load, though that seems needlessly artificial)
It's pretty easy to get inverters attached to a battery to act like "virtual inertia", in fact this kind of service to the grid is an area where they massively outcompete traditional turbines (providing far better frequency stability for lower cost). This concern is a complete non-issue if you have even a small fraction of the storage that a renewable grid needs.
It does both (while there's still spinning generators on the grid, those are exactly the same thing because of how the generators work. When there aren't any on the grid, they are in principle seperate, but the inverters will still use the frequency as a signalling mechanism, since it's a very robust means of controlling a system as large and complex as a grid).
I wonder if at some point, with solar and batteries on most houses, we'll make the switch to DC power inside the house. Maybe with some low voltage wiring (lights and such) and high voltage wiring circuits (other appliances).
This way we avoid the loss of converting from DC to AC and remove most of these issues entirely. Sure, a lot of appliances wouldn't work in DC power, specially appliances that have motors in them, so for those we can have the DC to AC unit.
Same thing for generating power back into the grid, just have one of these converters strapped to our house output/input.
It would seem that the frequency and synchronization could be solved by referencing both to the atomic clock, which is a time signal available most anywhere and via methods that are not grid dependent.
The problem isn't accuracy. It's not generally important to hold perfect frequency. The importance is in coordination. If the grid frequency lags a little bit, then other plants should also lag, to prevent being out of phase.
Also, I'm absolutely baffled why there's never been a civil defense initiative to get even a small baseline of solar/wind power at nearly every household. Even having just 1kW in every household would be enough to keep running basic communication, refrigeration, heating systems (i.e.. starters & blowers for gas/oil/etc. not elec furnaces), etc. In a major extended blackout situation, this would enable society to be sustainable for weeks to months or even indefinitely, instead of only days in winter as we have now.
Another huge concern is availability of transformers, which hafe a huge lead time. The attack on substations in Moore County NC [0] may be only a very successful test run. I've read reliable reports indicating that not many such attacks in the correct points on the grid could cripple the entire grid, and could also take out many transformers on the way down. Current production rates could take years to replace. Yet it would cost something like $500 million to stockpile them (shelf life is measured in decades), and AFAIK this has not been done. Why? Seems like just ordinary political oversight - it isn't sexy, even though it could be critical.
Not mentioned, you need to "harden" your electrics, if your power grid restarts frequently. Cause the spikes and surges fry weak electric devices. So its filters, timers and stability checkers to keep everything intact.
True, but it's not necessarily expensive either in the current context. The strike price EDF agreed for Hinkley Point C in 2013 was £92.50/MWh in 2012 prices. Wholesale prices (both day-ahead and next-season) have been well above that level for some time now (although it is also delayed and over budget so there is additional subsidy in the form of the pain-share mechanism to account for).
For new build baseload coming online in 10-20 years, coal or CCGT with carbon capture are probably the realistic alternatives to nuclear in most countries which don't have the resources to build out significant hydro or geothermal baseload. But it's not clear that carbon capture technology can be scaled enough (especially if we go down the storage rather than reuse route), and those technologies look like they could be equally as, if not more expensive than nuclear at this point.
Realistically we need to plan for nuclear to at least potentially be a significant part of the energy mix if we can't solve the current scalability issues with carbon capture, demand reduction and storage. On the plus side the costs per unit should scale down as we build more (as the French did in the 70s/80s), if there is the political commitment.
Any new start nuclear plant today would likely be more expensive than renewables + storage (to combine to produce equivalent sythentic baseload) by the time the nuclear plant could come online. I don't think there's any window left for new nuclear plants to make sense, unless their cost can be radically reduced.
I could be self sufficient for 10 months a year as a house with solar and batteries for about 10% of the cost of the house, and obviously would be able to opt out of consumption during high demand on the grid.
Yet we still allow new houses to be built without powerwalls or similar storage systems.
Numerous companies now offer all in one battery/inverter/management packs for something like $1 per Wh. Setting up a supplimentary system is several thousand dollars and could pay for itself in just a couple years. At this point, the hardest part of the process is finding a load you can power off your micro-solar-grid. Air conditioners work pretty well for that role.
Solar is now stupidly, mind numbingly cheap for what it is, and yet there are still people who think it "doesn't work" as if literal gigawatthours aren't generated daily by millions of people and systems.
Some people have Javascript disabled and/or third-party resources blocked by default, in which case the only indication that there is an embedded video (instead of just some kind of blocked advertisement) at the top of the page is the text "[Note that this article is a transcript of the video embedded above.]" at the top of the article. It's easy to miss.
(The first time I read an article on that site, it looked a bit strange to me, as if it were stage directions for a video; only by reading the comments here on HN I learned that it had an embedded video, and that the text was just a transcript of it. Unlike this one, that article didn't have a note at the top mentioning that it was a transcript.)
JavaScript is firmly established as how the web works. I don't think we need to bend over backwards for people who intentionally cripple their browsers.
As an engineer who has spent time commissioning electrical systems on oil rigs, I've had the opportunity to witness some truly fascinating tests. However, out of all the tests I've seen, the black start test is by far the most intriguing.
Similar to the article, but at a smaller scale, it takes power to start up one of the ~7MW main generators on an oil rig. During commissioning, a black start test simulates a scenario where all power sources on the rig are exhausted, and the emergency backup systems must be used to bring the rig back to life.
Starting the main generators on an oil rig is no easy feat. It takes a significant amount of power to get these massive machines running, and it's not as simple as pulling a cord like you would with a lawnmower. In normal circumstances, the rig uses tanks of compressed air to start the generators, but during a black start test, these tanks are assumed to be empty.
So, how do you start the main generators in this situation? The answer is with a special emergency hand-cranked air compressor. By cranking (and cranking, and cranking) this compressor, you can generate enough air to start a small air compressor, which in turn is used to pump up the air tank and start the 1.5MW emergency generator. Once the emergency generator is running, it can be used to power the compressors that fill the large tanks needed to start the main generators.
Watching this process unfold is truly a unique experience. To see a massive oil rig slowly come to life, all thanks to one person cranking away at a small air compressor, is truly impressive.
Unrelated but fun fact: some heavy duty air-brake trucks can also use compressed air to start the engine, but what if your tank is flat? You can hook up one of the tires and use that to start the engine! Really neat.
Also, many trucks have both an electric starter and an air starter. I one day aspire to those kind of redundancies in a vehicle I own.
Very much reminds me of the stories of my dad how they'd have to start really big bulldozers back in the day of draining german marshes. You'd first hand-crank a small generator to fill up a small battery for the electric starter motor. Once the battery is sufficiently full, you'd engage the starter motor, which in turn spins up a flywheel at higher torque than the crank could. Once that flywheel was up to a certain amount of speed and momentum, you'd engage that with the main crankshaft of the big old engine and hope it had enough power to push the big engine through one or two strokes. That's what it needed to start running on its own. Otherwise you'd have to start over.
Big machines and complex startup machines are surprisingly fascinating.
I don't get why you need to hand pump a bunch of air just to start a small generator/air compressor when I can pull a rip cord and have 10kw of power in seconds and use that to bootstrap the remainder of this startup process.
It would seem you only need to do a bunch of pumping to startup a larger engine where the smaller starter/pony motor is not working. I wouldn't describe such an engine as 'small' though.
So if I'm understanding the description above correctly, the chain is:
turn crank -> bootstrapping air compressor -> air tank 1 -> 1.5MW emergency generator -> bigger air compressor -> air tank 2 -> 7MW main generators
Pretty rube goldbergian.
Here's an interesting parallel: the process that a computer goes through when it booted is itself an incredibly complex, intricate, and rube goldbergian process. Unlike a power station black start, this process happens completely automatically every day when someone turns a computer on, yet it is no less rube goldbergian and in many ways probably more so.
The boot processes of computers have always fascinated me for this reason, and I suspect black start processes fascinate me for much the same reason. It's not unlike why people enjoy watching videos of rube goldbergian contraptions.
It is similar how planes starts. First small turbine engine (APU) starts. When APU fails to start, they use specialized airport cars with compressed air. Then the turbine helps to start first main engine. Then the main engine helps to start second main engine. Repeat until all main engines are started.
(https://www.youtube.com/watch?v=GzhdxSsoT0g)
This is one of the services that interconnectors provide to the grid in the UK. Should a black start be needed and Europe can still provide power then an interconnector can be used for black start.
Only skimmed thru, but seems to neglect to talk about the capacitive charging needed to energise the transmission network, grid where I live is isolated and needs 25MW plus just to bring the transmission network up.
Amusingly, I encountered the same sort of problem on a vastly smaller scale: An electric car with a fully charged 400 V, 64 kWh battery that I could not start for lack of electric power.
Said electric power normally to be delivered by a small conventional 12 V car battery that was, unfortunately, almost completely discharged. To get my car started, that battery has to provide power to some electronics and to close a relay connecting the 400 V battery to a power inverter providing power to the car – including power to charge that 12 V battery.
The owner of a gas car that I asked to help me jumpstart my car would not believe me at first, but I got the help I needed, thus saving the Christmas party I was going to on that occassion.
105 comments
[ 3.3 ms ] story [ 177 ms ] threadFurthermore, some mods make it so that even pumping the water to the boilers requires power (*), and add a separate "burner generator" that directly burns coal to produce power. So you end up setting the automation so that tier one power generation uses the burner generators to build up water reserves, then tier two power generation turns off the burner generators and uses the boilers to build up steam reserves, and then finally switches on the rest of the factory once steam reserves have been built up.
(*) In the base game the water pumps magically work without needing electric power, because the devs don't want to complicate the experience of setting up power. So it's an amusing situation where the Factorio character starts with the innate knowledge of how to build magic water pumps and steam engines out of iron plates and gears, but needs to "research" more trivial things like dumping rocks into a lake to landfill it and create a buildable surface.
We were taught the basic physics of the Newcomen engine in school as part of history on the industrial revolution, but nothing about geo-engineering for land reclamation.
the steam engine is the only (early) way to produce electricity, and the pump needs electric, and the steam engine needss water
(I tried googling for the mod but came up short)
https://mods.factorio.com/mod/aai-industry
The bigger magic: hand crafted belts work for transporting goods reliably over long distance without any energy supply. Quite some engineering there.
You did need to power the water pumps and it sucked depending on the distance.
Likewise, magic belts because belts come far earlier than power.
Later, I move to nuclear power, where fuel use is such that even glacial low-power inserters can keep the facilities fed through a power spiral. Although my massive 12-reactor facility did rely on pumps to keep the steam moving fast enough through my flowmeter to regulate power... but when you have a 12 reactor powerplant, power death spirals take long enough to occur that you will have ample time to fix them before things start needing black start.
I do agree that if you try to go for a most optimum setup you do risk getting in a black start situation.
My current game I'm trying a totally different idea--I'm trying to play it with no defenses at all, rely purely on killing biter bases before they get upset.
Unfortunately I just discovered that they removed it from their website somewhere this year. I've asked them to restore it. For now you can play with this copy in the Internet Archive: https://web.archive.org/web/20220124003709/https://www.entso...
And since it is never tested, it's almost guaranteed to fail in some unexpected way.
And since that's the case, almost anything is better then getting yourself into a black start scenario, including shedding load in any possible way.
Which leads to it never being tested, and more things being unknowingly broken.
For example, the entire station may be disconnected from the grid, but somewhere in the system is a Siemens device that for whatever ungodly reason needs to access the internet, which works when the grid as a whole is operational but fails when it can't get cell service.
And even if those stations work correctly, you still haven't verified that the actual plan is operational because you haven't actually done the black start; perhaps the grid divisions are incorrect and the first load you try to apply will brown out the start source.
It's the same reason that datacenters don't just run the generators every month, but actually switch over to generator power and disconnect from the grid to verify it works (and if it doesn't, hopefully the batteries last long enough to get back to the grid!).
Looks like they do test the procedure.
>To be able to restart the plant, it made use of versions of jet engines based on those used on Concorde to jump start the main generators
https://en.wikipedia.org/wiki/Cruachan_Power_Station
Edit: Given that the turbine hall is 300m deep in granite that station is probably fairly safe from anything other than a nuke!
300m deep in granite is safe from a nuke as well.
It's built inside a hollowed out mountain - definitely worth doing the tour if you're in the area.
Edit: apparently Drax as well as many of the older plants can too.
"The power station was also promoted as a tourist attraction, with visitors able to take a minibus trip from "Electric Mountain" - the name of its nearby visitor centre - to see the workings inside the power station;[18][17] 132,000 people visited the attraction in 2015.[19] However, the centre is now closed with no prospect of reopening."
Isn't that the sorta thing you would rather not let any others know, you know, for the sake of, national ... you know?
And if it is known in the industry it is absolutely guaranteed to be known by all countries, friend and foe alike; and real "foes" will have even more knowledge.
What is kept secret is what tooling the various militaries may have to assist in a black start, and what would trigger them actually attempting assistance.
Another example is blueprints - they seem very detailed and accurate until you realize that quite a bit is "left out" and the tradesmen are assumed able to figure it out - it will show fixture locations for example, but rarely show exactly how the wires are run to them.
But if you're creating a 'map' of a house you bought, wire locations is extremely useful because they're all behind the walls.
Theoretically solar panels in a good place to do it: can output power without needing input power, can have several megawatts on a site, can match phase very easily, can potentially generate phase.
Admittedly you have to wait until the sun comes up, but in a true black start situation things are going to move slowly.
Starting back up is complex because loads need to be added to the network in some order they can be powered.
Powering them down to a complete stop is a rather long process taking hours.
One silly technician left a spanner in one turbine after maintenance, and it took about 2 years for a replacement part to be installed.
Almost all plants will trip from the grid if it gets bad and remain idling for some time. But if that happens, the grid is still dark and the black start generators are the best equipped to bootstrap the grid. You can't start with the big generators too, so you can initially only bring in small plants until the grid capacity is big enough to not trip out when the big generators come online in a wrong way.
There is a good chance that by the time the grid is stable, atleast some power plants will have had to shutdown entirely and spin back up from a black start with only external power.
We have a few cycle gas turbines (diesel) and two major hydro schemes which maintains a critical reserve at all times of water, so they just need to open the sluices to jumpstart the system.
Also, our grid is tied to our neighbors so they can help out if need be like France does to the UK during shortfalls.
I will have no power from 6pm until 8:30pm and then again from 10pm until midnight.
Fortunately, we got a gas cooker/inverter for my ONT/Router/WiFi.
Does the ISP keep things working for the duration?
The core routers in probably in a datacenter with battery/diesel backup.
We had outages in the beginning because they use DHCP for the fiber box to get an IP address so you can imagine a few thousand ONT terminals all coming online nearly at the same time and hitting one DHCP server
The grid is “decentralized” but still brittle and at risk of cascading failures. A future where the majority of power is provided by onsite solar + storage is much more robust.
Similarly not robust are things like tech stacks with complex “cold boot” plans or unexpected cascading failure modes, etc.
I don’t know what the right terminology for this subset of decentralization is.
It's cheaper to provide additional demand via fossil fuels than to rely on storage solutions. You need a huge amount of storage.
Intermittent scarcity and over supply of energy (both are a thing) also leads to different consumption patterns. People valuing cheap energy when it is scarce invest in batteries that they charge when there's a surplus. Others buy at a premium or simply shift when they consume their energy. Demand shaping is also something that can be incentivized financially.
Of course that works great in Australia but other parts of the world have other options like wind, geothermal, hydro, nuclear, etc. Which is why places like Norway and parts of Canada are also mostly running on clean energy at this point.
The key is actually not isolating but interconnecting. Norway is part of a connected grid that spans most of Europe with a market where you can sell energy and buy energy. Norway is an energy exporter at this point and there are plenty of countries importing their power to supplement their own generation when it falls short. That's why countries like Germany are now predominantly wind and solar powered and yet doesn't have blackouts when there's no wind. Like today (gloomy winter day with barely any wind here in Berlin). But there's still plenty of wind elsewhere. Germany imports almost as much power as that it exports each year.
Balancing the grid is hence a matter of 'maintaining frequency'. Which makes the frequency very stable.
Solar and batteries are inherently DC. They have converters that take a reference AC signal, and output their energy synchronized to that reference. Incidentally, this makes it difficult to run your house off solar during a black-out, because you lack the reference. Even in smarter converters that don't need a reference to match the frequency, they still need a reference to synchronize the phase.
When we move to more batteries and solar for generation, that will put more strain on the spinning-based generation for short-term fluctuations. Less spinning mass means less buffer. At some point, any flaws in behavior of DC converters might start dominating. If they happen to resonate with each other, we might see some weird things.
It is true that if you insist on making the cutover from inverter derived AC to grid AC without an interruption, you will need some subtle circuitry that can automatically "glide" the inverter into synchrony with the grid before cutting over.
I don't know if Tesla, Bluetti, and EcoFlow devices do this, or just make a step-change in the phase and frequency of the output. That would be an interesting investigation, maybe Matthias Wander can do it?
If you are out of phase when the grid gets back, it gets damaging. So it is difficult to get an approved system that can do it.
You could try to drag the grid to a perfect frequency by lagging or leading the pattern. But that feels like it wouldn't work.
This way we avoid the loss of converting from DC to AC and remove most of these issues entirely. Sure, a lot of appliances wouldn't work in DC power, specially appliances that have motors in them, so for those we can have the DC to AC unit.
Same thing for generating power back into the grid, just have one of these converters strapped to our house output/input.
Perhaps it is not accurate enough.
Also, I'm absolutely baffled why there's never been a civil defense initiative to get even a small baseline of solar/wind power at nearly every household. Even having just 1kW in every household would be enough to keep running basic communication, refrigeration, heating systems (i.e.. starters & blowers for gas/oil/etc. not elec furnaces), etc. In a major extended blackout situation, this would enable society to be sustainable for weeks to months or even indefinitely, instead of only days in winter as we have now.
Another huge concern is availability of transformers, which hafe a huge lead time. The attack on substations in Moore County NC [0] may be only a very successful test run. I've read reliable reports indicating that not many such attacks in the correct points on the grid could cripple the entire grid, and could also take out many transformers on the way down. Current production rates could take years to replace. Yet it would cost something like $500 million to stockpile them (shelf life is measured in decades), and AFAIK this has not been done. Why? Seems like just ordinary political oversight - it isn't sexy, even though it could be critical.
[0] https://www.npr.org/2022/12/05/1140775417/north-carolina-sub...
South africa has made some experiences there.
When I am talking about batteries, I really talked about NBD style devices[1].
https://ndb.technology/
For new build baseload coming online in 10-20 years, coal or CCGT with carbon capture are probably the realistic alternatives to nuclear in most countries which don't have the resources to build out significant hydro or geothermal baseload. But it's not clear that carbon capture technology can be scaled enough (especially if we go down the storage rather than reuse route), and those technologies look like they could be equally as, if not more expensive than nuclear at this point.
Realistically we need to plan for nuclear to at least potentially be a significant part of the energy mix if we can't solve the current scalability issues with carbon capture, demand reduction and storage. On the plus side the costs per unit should scale down as we build more (as the French did in the 70s/80s), if there is the political commitment.
Yet we still allow new houses to be built without powerwalls or similar storage systems.
Solar is now stupidly, mind numbingly cheap for what it is, and yet there are still people who think it "doesn't work" as if literal gigawatthours aren't generated daily by millions of people and systems.
If everyone does that, how much do you think the grid is going to cost you for those other two months?
(The first time I read an article on that site, it looked a bit strange to me, as if it were stage directions for a video; only by reading the comments here on HN I learned that it had an embedded video, and that the text was just a transcript of it. Unlike this one, that article didn't have a note at the top mentioning that it was a transcript.)
I know that.
JavaScript is firmly established as how the web works. I don't think we need to bend over backwards for people who intentionally cripple their browsers.
Similar to the article, but at a smaller scale, it takes power to start up one of the ~7MW main generators on an oil rig. During commissioning, a black start test simulates a scenario where all power sources on the rig are exhausted, and the emergency backup systems must be used to bring the rig back to life.
Starting the main generators on an oil rig is no easy feat. It takes a significant amount of power to get these massive machines running, and it's not as simple as pulling a cord like you would with a lawnmower. In normal circumstances, the rig uses tanks of compressed air to start the generators, but during a black start test, these tanks are assumed to be empty.
So, how do you start the main generators in this situation? The answer is with a special emergency hand-cranked air compressor. By cranking (and cranking, and cranking) this compressor, you can generate enough air to start a small air compressor, which in turn is used to pump up the air tank and start the 1.5MW emergency generator. Once the emergency generator is running, it can be used to power the compressors that fill the large tanks needed to start the main generators.
Watching this process unfold is truly a unique experience. To see a massive oil rig slowly come to life, all thanks to one person cranking away at a small air compressor, is truly impressive.
Also, many trucks have both an electric starter and an air starter. I one day aspire to those kind of redundancies in a vehicle I own.
Big machines and complex startup machines are surprisingly fascinating.
It would seem you only need to do a bunch of pumping to startup a larger engine where the smaller starter/pony motor is not working. I wouldn't describe such an engine as 'small' though.
Here's an interesting parallel: the process that a computer goes through when it booted is itself an incredibly complex, intricate, and rube goldbergian process. Unlike a power station black start, this process happens completely automatically every day when someone turns a computer on, yet it is no less rube goldbergian and in many ways probably more so.
The boot processes of computers have always fascinated me for this reason, and I suspect black start processes fascinate me for much the same reason. It's not unlike why people enjoy watching videos of rube goldbergian contraptions.
Said electric power normally to be delivered by a small conventional 12 V car battery that was, unfortunately, almost completely discharged. To get my car started, that battery has to provide power to some electronics and to close a relay connecting the 400 V battery to a power inverter providing power to the car – including power to charge that 12 V battery.
The owner of a gas car that I asked to help me jumpstart my car would not believe me at first, but I got the help I needed, thus saving the Christmas party I was going to on that occassion.