Load drops, the big spinny generators stop having so much resistance and accelerate.
Load increases and the big spinny generators have more load, synchronously slowing down.
So in simplest system just feeding your generator more when frequency is below nominal and less when it is above is enough
I'm glad there's at least two of us ;)
"What happens when I turn on an electric device in my house?"
"A turbine in a power plant spins more slowly for just a moment."
By far and away the easiest way to control production - i.e. brake/feather turbines or dump PV, is frequency shifting, as I’ve batteries and inverters in several locations and while networking them would be possible, it’s unnecessary. It’s typically intended for grid-tie operations, but here I use it to control our tiny isolated grid.
It’s a pretty small range (50.2-53hz) over which they shift, but it’s more than enough.
Practical engineering has a great video about power black starts that give some insight into this complicated machine.
PS One of the answers in the SO thread mentions JET in the UK. I spent a few summers there as an electrical engineering student (it's home to the MAST and JET fusion reactors). When the JET tokamak ignites a plasma, it can't sustain it for very long (we are not yet at the point of extracting enough energy to sustain the reaction). As a result they need to ignite the plasma and keep it hot. They can't do it for more than 1-10 seconds. During that time, they draw massive amounts of power - they're permitted to draw up to 1% of the UK's capacity for a short period, whilst they simultaneously dump all the energy stored in two gigantic flywheel generators housed in a nearby building. I've never been there when the flywheels are running but I've climbed around beneath them. There's nothing quite like massive engineering :)
Generator spins at 50hz Motor spins at 50hz
Add load to the motor, it starts slowing down both itself and the generator. Generator governor increases input energy, frequency goes up. Both are in phase all the time.
Remove load, both start spinning faster. Reduce governor to regulate it to 50hz again.
It's a bit harder with inverters but the idea is similar, you follow the grid phase and if you want to send energy to the grid you will be slightly early to the grid phase and if you want to take energy from the grid your phase will slightly lag.
The difficulty is not "keeping them in phase", that just happens (aside from initial connection), it's the whole load prediction and handling, when to tell which plant to start producing more or less energy, with variety of plants having shorter or longer ramp-up/down periods
The most difficult bit for an operator is to make sure the generator is synchronized before they actually connect it. Only after it's synchronized can they start actually feeding power in.
You can't just turn these things on and off at will like a regular motor. To extend the gears analogy, gears need to be synchronised before they can engage - just like synchronous AC generators.
[0] https://arstechnica.com/tech-policy/2021/05/texas-power-outa...
Battery production is currently shifting from hundreds of gwh per year to twh per year worth of production. They last quite long too. If you cycle 1 twh of battery every day, it would run out in about a decade (assuming 3000 cycles or so). And if you do that, that amounts to absorbing and serving about 365 twh.
The worlds entire electricity production is measured in peta watt hour (pwh). About 25 phw. or 25000 twh. So, with battery production scaling int he tens of twh per year, there may soon be enough battery to absorb and produce that kind of volume. Doesn't mean it will be used exclusively for that. But it drives the point home that we are talking some very non trivial levels of standby power. That's car batteries, home batteries, grid batteries, etc. connected to the grid. Mostly neither charging or discharging most of the time.
Tesla and a few others are starting to experiment with virtual energy providers. That might go from a few GW to having in the order of TW of power available on demand very shortly. There are about half a million tesla home batteries out there apparently. Even a few thousand of those already add up to quite a bit of collective power. Imagine that scaling to millions of connected batteries. They'd be the largest power supplier, by far, in most markets.
That won't run the grid for hours/days obviously. But it does eliminate a lot stand by capacity that is expensive and poorly utilized.
That's also the future for energy intensive industries. They'll want to generate most of their power locally (because that's cheaper) and use batteries to buffer it and fall back to the grid only when that runs out and export to the grid whenever there's a surplus locally. So, that reduces load on the grid and evens out a lot of the load spikes. And a lot of the time they'll be feeding energy into the grid. Anything energy intensive is going to compete on cost. And that means investing in local production and batteries.
Consider how many obscure but useful, necessary even, things we've learned over the past thousand years.
Before the internet it would all be hidden away, only quickly available to specific experts in their fields.
You'd have to go to a university library and dig for hours.
It's actually a minor plot point in Stephen King's "The Stand" that after a viral apocalypse wipes out most of humanity, the survivors try to restart the electrical grid in a town.
The first attempt proves fatal because they haven't properly isolated the circuits throughout town, so when they connect the plant to the main grid the plant's generators explode from trying to support the load of every home where someone died spontaneously while running a hair dryer.
https://pt.wikipedia.org/wiki/Museu_da_Eletricidade_(Lisboa)
But to really know something, you have to have studied it, even in the modern era. I can read an instantly-available wikipedia article about power generation, but I'm only scraping the absolute surface.
If it's not online it doesn't exist. /s We really need more of these obscure pieces of knowledge put online because people don't go to the library as much as they used to.
There's so much specialized knowledge out there, but often you'll only learn about it from someone else, or a highly technical book.
Like, say you want to frame and build your own house - there's bazzilions of YouTube videos about the beginning process. But very very few about the more advanced details you need to know.
That's an interesting factor of online knowledge - the most readily available one is the one that masses are interested the most because the views give the budget for people to care. It's also weirdy trendly, like how pandemic spawned a lot of woodworking channels coz people cooped up in their homes found a new hobby.
To add to the point this [1] random video is an example, it does go into terminology and reasoning behind each element but won't tell you what kind of lumber to get, how climate would affect that decision, how to isolate the house and a bunch of other things. You might look for them and probably find some info but at some point just looking for a dedicated book might be the saner option.
And uh, if someone knows a good one covering how to build and isolate your own shed I wouldn't mind recommendation...
I remember my mind being blown by the Cal engineering library. Endless stacks and stacks of dense foundational knowledge from the 1950s-1990s you would never find on Google. Books are massively underrated these days.
Even if it is online, it probably doesn't exist either - see today's threads about communities moving from Reddit to Discord. The "cozy web" is undoing a lot of progress of the past decades. Might be that we'll all need to go to the university libraries and dig for knowledge ourselves, hoping any of the more recent experiences and discoveries end up being published as books, instead of dying in private Whatsapp groups.
There is so much deep information available in any large university library that is simply not on the internet at all. If you're researching any historical topic, there are at least a few solid books (and possibly hundreds) that draw from primary sources, compared to a couple pages of text on Wikipedia and very little else. You mostly won't even find ebooks.
Somehow the best, most extensive free digital resource is a podcast like Age of Napoleon, which synthesizes information from many books.
Advanced? All I need is some TooBa Fours and a Larry Haun book...
On one of my first IT jobs at a big manufacturing company my team was tasked to find out why there are regular power outages in some printer rooms (there were rooms with shared printers on each floor of the office building). There were always some tripped circuit breakers and the facility management had to dispatch someone to put them back on. Between those incidents were always some weeks were nothing happened, but when it happened it affected a lot of printer rooms.
In the end we found a monthly cronjob on a central printing server which triggered a testpage print on all connected printers. Took us quite some time since no one ever saw those test pages. Never underestimate the needed current for a room full of colour laser printers coming to live all at once.
Part of the printing process involves passing the paper covered in toner through a hot roller, to fuse (melt) the powder toner (ink) onto the page. That roller has to be up to temperature to print. It is normally heated by a powerful (ie. 1 kilowatt) light bulb inside a hollow roller. Sometimes if you peek through the vents in the printer, you can actually see the light it makes.
The light bulb is pulsed on and off to maintain the right temperature - but when coming out of standby it is solidly on for ~10 seconds. Manufacturers want their printers to warm up from standby quickly, so they put very powerful heaters in them, even though the steady state heat requirement isn't awfully much while printing.
Also: don't put a laser printer in your bedroom. It's unhealthy. Only learned about that after the fact.
3D printers are even worse, depending on the filament type (ABS is worst?). Always ventilate!
A few papers printed over the course of years won't kill you. What will are the conditions of working adjacent to the office copier, 8-10 hours a day, for years.
Get an air purifier to capture particulates. (Supposedly, houseplants help too.)
The drive array was powered by 6x, 400W ATX server supplies with my own wiring harness. This was enough to keep them running but they had to be sequenced carefully to keep from overdrawing the power supplies.
This was all on an UltraSPARC 6k so there was plenty of support for that; bringing up the system always sounded like multiple jet takeoffs tho. Took 15min. When the rack of 10k RPM "quick cache" disks spun up it was like a chorus of the whines of the damned.
I'm going from memory here but each Ultra 320 SCSI HDD had a startup current of almost 2 Amps so if you had a disk shelf with 24 drives and stack a few shelves in each rack you could do some serious power damage if you didn't plan the startup sequence right.
New Zealand had to build an aluminum plant in the 1960s and figured out it would be more efficient to bore miles underground, build turbines, install a huge power station, and wire it tens of miles to the smelting plant. It relies on a vertical drop between Lake Manapōuri and the open ocean to create the gravitational potential to turn the turbines. When the smelting station doesn't use the full capacity the power station has to immediately reduce output, otherwise it can overload the transmission lines to the rest of the grid.
Manapouri Power Station is the name if anyone is interested in reading more. It has an interesting history. https://en.wikipedia.org/wiki/Manapouri_Power_Station
They are at it now. That place closing down would lose 1000+ jobs if n a region that doesn’t have a ton of options, the world would lose a greenish supply of aluminium and I would potentially get a discount on my power bill when I stop subsidising that massive company (that smelter uses a bit over 10% of NZs electricity).
Oh, and Rio Tinto have dumped toxic waste in various places too.
https://www.powercompare.co.nz/n/tiwai-point-aluminium-closu...
https://www.rnz.co.nz/news/national/468066/tiwai-point-toxic...
My brain however has in the past created incredible nightmares based on the scale of what I experienced there. I just _had_ to see what the tailrace output looked like and was not disappointed by this nightmare fuel: https://www.youtube.com/watch?v=qlhSzs4JXE0
The longer route (which actually flows next to the farm my mum grew up on) doesn't even go near a major population centre. The largest town (Tuatapere) has a population of 500 people.
Several dams in the highlands create multiple reservoirs, the water is then fed through kilometres of tunnels before it reaches a ~400m vertical penstock to the Fljótsdalur underground power station (the power station is about 40km / 25mi inland, while the smelter is on the coast)
So the entire grid is in phase, by which I mean every generator, load, &c is operating on the same AC pulse (handwave-handwave ignoring smaller loads, transformations, etc.). But that phase isn't instantaneous; it's near-light, and the grid is long enough for that to matter in places.
So every point in the grid can't be perfectly in phase with every other point, right? Because we have both lightspeed delays and loops, so even if point A is receiving power from two substations in-phase, point B (with different lengths of wire to those two substations) should be receiving it out-of-phase, right? How do we balance that in the grid?
That’s described on https://en.wikipedia.org/wiki/Synchronization_(alternating_c....
Firstly, you design your generator to have the same wave form and phase sequence as the grid.
Then, you power up your generator and make sure the voltage, frequency and phase angle of the electricity it generates matches that of the grid.
The moment that’s (more or less) the case, you connect the grids, preferably on a zero crossing. That Wikipedia page describes a setup with incandescent bulbs that can be used to manually detect that moment, but nowadays, it’s done using electronics.
Maybe imagine a line of buoys in the ocean as a wave passes.
(Perhaps the obvious answer is correct here: "Grids don't work that way; you'd never do that.")
You'd have to have 2 different routes from same power source that differed in hundreds of kilometers in length and that just doesn't really happen. And small phase difference would just cause uneven loading
Also that power would be "used up" closer to the power source and you'd be sucking off power from closer sources.
As far as I know, in your example, one of the two circuits will be transmitting more power from the substations to point B than the other. There's a device called a phase-shifting transformer (https://en.wikipedia.org/wiki/Phase-shifting_transformer) which can be used to adjust the phase angle of the circuits, and that way, adjust how much of the power is carried by each circuit.
(The following sentence from that article probably goes to the core of the answer for your question: "For an alternating current transmission line, power flow through the line is proportional to the sine of the difference in the phase angle of the voltage between the transmitting end and the receiving end of the line.[1]")
If everything is in a line, you just sync to your local phase, but if you have wires in a triangle that's impossible, and they basically just ignore it, in practice it doesn't happen often enough to cause any problems.
So the grid doesn't need to balance it, no.
And from the receiving end's perspective, power will be exactly one phase out of sync if the cables differ in length by 5000 km. I'm assuming that isn't really a problem people have to worry about, but my background is in physics and not engineering (i.e. maybe there's some industrial applications that I'm not thinking of).
Everyone isn't always in phase with every other station's phase at that exact moment in time. They're just in phase with their local part of the grid's phase. Though amount that is different is milliseconds, not seconds as in the stadium wave example.
This mostly came from reading this: https://electronics.stackexchange.com/a/291328
not in phase as such, but in sync.
because you only have to match your 60hz(or 50hz) to what you receive locally, so long as the waveform is reasonably stable its not that much of an issue.
Because everything is effectively a bunch of elastically linked pendulums, you tend to reach equilibrium, so long as the load doesn't act too much as a dampener
[1] https://en.wikipedia.org/wiki/High-voltage_direct_current#Ba...
Before starting up the fans, the guy running the control booth picked up the phone and had a short conversation, roughly
"Hi, this is [name] at the wind tunnel; can we turn it on?"
[someone on the other end replies]
"Great, thanks."
I asked who he was calling, and he explained that he had to check with the power company before powering it on. This was mid-winter, so grid demand was low; apparently during the summer (when everyone has ACs on), the start-up load could cause brownouts!
https://en.wikipedia.org/wiki/Shearon_Harris_Nuclear_Power_P...
"The Shearon Harris site was originally designed for four reactors (and still has the space available for them), but cancellation of an aluminum smelter plant in eastern North Carolina in the 1970s resulted in three of the reactors being canceled."
There was so much current that the free-air magnetic field in the plant was literally palpable, the engineer giving us the tour did a demonstration similar to this video[1] which blew my young mind.
A portion of the land, and the substantial power-handling infrastructure (and proximity to the river for cooling) now powers a Google datacenter.
That means that each MW/hr of production only makes 18 kg/hr, so a 900MW/hr nuclear plant only makes 8 tons of aluminum a hour. Its insane.
Secondly, you're not wrong, the way that aluminum is smelted is by melting e.g. bauxite or another aluminum compound, and then electrolyzing the resulting fluid to extract pure aluminum. Usually the same electrodes are used for both operations. It's the very grandest scale of electrochemistry, and the reason that aluminum smelting plants are nearly universally located near cheap and highly available power sources.
* Something close to 900Mwh, anyway, given that reactor nameplate capacity is not always the actual running power or peak possible output, plus an allowance for maintenance. Other power sources have different capacity factors that would result in something below 900Mwh, but a typical fission plant is "up" continuously for our purposes
There aren't many places where electricity can be produced close to that. Iceland is one and it's unsurprisingly the world's major bauxite importer and aluminium exporter. Wholesale electricity there goes for around $42/MWh [1].
OK, the smelters can do a bit better than the average wholesale rate, but not much - Iceland's electricity supply is not highly variable like solar or wind.
So the rest of the expenses of the process - mining, shipping half way around the world twice, capital costs - are all basically free compared to the electricity cost.
[0] https://markets.businessinsider.com/commodities/aluminum-pri...
[1] https://www.datacenterdynamics.com/en/opinions/time-for-an-i...
---
We built a whole application to manage a certain workflow between multiple systems, and our users kept using email instead, because it turns out that the users on the other end don't check the system, but they do check their email.
You also have people who are drilled in following procedures, especially when failures in process become very public.
Systems had been slowly added over the years, but because the power system is pretty reliable around here, they'd never had to start things up from a complete power failure.
Big power consumers usually pay for power at grid market rates, which vary from hour to hour. So they're tied into both the market system and the control system. This is done via a Curtailment Service Provider.[2] Some of those are power distribution companies, and others are just brokers.
Here's one in California.[3] There's a phone app, a web page, a connection to your meter, and an API for your own load's control system. Large power consumers connect to them, and they connect to the grid operator, which is CAISO for California. Once everything is connected, they can remotely tell your systems to reduce their load and verify that has happened, for which you get a price break.
There's the Peak Load Management Association, which you can join if you buy power by the gigawatt.[4]
[1] https://pjm.com/-/media/about-pjm/newsroom/fact-sheets/deman...
[2] https://www.pjm.com/markets-and-operations/demand-response/c...
[3] https://cpowerenergy.com/wp-content/uploads/2018/01/CAISO_DR...
Does that mean that the 775 ton flywheels could come to a complete stop in less than 13 seconds if fully used up?
A heavy contactor powers on stage 1, stage 1 powers another heavy contactor powering stage 2, stage 2 does the same with stage 3.
When mains voltage is too low, the contactor cannot close. This way this allows for tens of milliseconds of separation in between power ons.
