From NOT A LOT OF PEOPLE KNOW THAT
By Paul Homewood
Layman’s guide to system inertia from Kathryn Porter
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
From NOT A LOT OF PEOPLE KNOW THAT
By Paul Homewood
Layman’s guide to system inertia from Kathryn Porter
Subscribe to get the latest posts sent to your email.
Notwithstanding Ms Porters MSc Physics qualification there is a fundamental mistake in the whole analysis. Electricity is NOT energy. Her ‘inertia’ is the energy stored in rotating masses – Millibands flywheels. Any consumer of energy delivered by the electrical grid has to get the energy from somewhere. You switch on your kettle and this immediately draws energy from – where?.
Wind and solar rely heavily on electronics to match the energy input to the grid status, but electronics do not store energy for very long.
The problem of demand spikes has been recognised for a long time, with switched loads operating at different times.
Renewables do not provide any significant stored energy, and indeed at night or for a windless period, any energy at all.
The current solution, embodied in the recent Electricity Bill ( would need to check exact details) recognises the necessity to control the load and mandates disconnectable smart meters. Few people seem to have noted this. EV chargers have to have demand control.
These are all details which are going to have a serious impact and ultimately make the use of renewables uneconomic. Perhaps the recent nuclear U-turn is a good sign.
I don’t think that was her main omission, though it was poorly explained. Inertia buys time, to increase generation or, if unavoidable, shed load in an orderly way. Her main omission was the newer methods of grid-forming inverters (on W&S, or any source of DC power). Flywheels have very good short term response, but don’t store much energy, so you don’t buy much time. Grid forming inverters backed by batteries have a lot more stored energy, and so buy a lot more time, and so stability. Of course, in the end, the only solution is more generation or less load. But stored energy helps you get there.
The only solution, Nick, is to ditch the virtue signalling green bolleaux and go with what really works: Coal, Gas, Nuclear and Hydro – where available.
The Physics Behind the Spanish Blackout
https://www.windtaskforce.org/profiles/blogs/the-physics-behind-the-spanish-blackout
By Bjorn Lomborg
.
Madrid knew solar and wind power were unreliable but pressed ahead anyway.
.
As I wrote in these pages in January, the data have long shown that environmentalists’ vision of cheap, reliable solar and wind energy is a mirage.
The International Energy Agency’s latest cost data continue to underscore this: Consumers and businesses in countries with almost no solar and wind paid, on average, 11 cents/kWh in 2023; but at 10% W/S costs increase to 15 c/kWh; at 20% W/S, 19 c/kWh; at 30%, 23 c/kWh, etc.
These numbers surely are an understatement
Germany pays 40 cents, nearly 3 times the US rate and nearly 4 times China’s.
No wonder it has become uncompetitive in world markets.
Adding tens of millions of illegal, unskilled, not-integrated misfits, from all-over is no help;
No help in the US, France and the UK as well.
Spending 3 to 4 times more on defense is no help either
The Spanish grid and generation mix will require expensive upgrades/changes to ensure the standard 99.98% reliable delivery of electricity, unless Spain—and its neighbors—are comfortable with increased frequency and severity of brownouts/blackouts.
Addition
Any energy systems analyst would know Spain/Portugal-like blackout problems would eventually happen, before a single W/S system were connected to the grid, but naive, woke, technically illiterate enviros do not want to listen to the pros. All wind/solar/battery nonsense must be stopped dead by taking away the generous subsidies.
.
The More W/S Electricity on the Grid, the Less the CO2 Reduction/kWh, due to Inefficiencies
Analysis of 2013 data of the island Irish grid showed the CCGT fleet operating at about 50% without wind; at 45.58%, with 17% wind.
At higher W%, the CCGT fleet operates at lesser efficiencies (high Btu/kWh, high CO2/kWh), until no CO2 is reduced.
Fortunately, Brussels paid for major connections to the much larger UK and French grids.
As a result, most of the ups and downs of wind output disappeared in the noise of the large grids.
https://www.windtaskforce.org/profiles/blogs/fuel-and-co2-reductions-due-to-wind-energy-less-than-claimed
.
Expensive Wind/Solar Systems
The over-taxed, over-regulated taxpayers and ratepayers are paying at very high rates, c/kWh, for: 1) electricity, and 2) Heat Pump heating/cooling, and 3) EV driving.
There is no way such high-cost electricity would increase standards of living and increase the GDP.
Businesses and skilled people would move to low-energy-cost states.
These businesses and people are tired of paying for:
.
1) Highly subsidized, expensive W/S systems that disturb the grid with weather-dependent, variable, intermittent electricity, which has caused expensive brownouts/blackouts, as in Spain/Portugal, California, Texas, New England, etc., and many other places, over the years.
2) Grid expansion to connect all these far-flung wind/solar systems to the grid,
3) Grid reinforcements to ensure the grids do not crash during periods with higher levels of W/S power
.
Here are some operational realities of W/S systems that are at the core of their problems:
.
Synchronous Rotational Inertia, SRI, Stabilizes the Grid
Closing down traditional plants (nuclear, gas, coal, hydro), with rotating generators that provide SRI, de-stabilizes the grid; a death sentence for the grid.
.
Wind/solar systems provide ZERO SRI to help stabilize the grid, because their variable outputs are digitized, then reconstituted into an artificial sine wave with the same phase and frequency as the grid.
Super expensive battery systems provide ZERO SRI.
Battery systems can provide virtual inertia, at very high c/kWh, by means of their back-end DC to AC power electronics (which failed in Spain/Portugal), which can quickly counteract voltage/frequency drops for a short time.
.
Connections Between Grids
Almost all grids have connections to other grids for import and export purposes.
About 50% of such connections are high-voltage, direct-current lines, HVDC
Such DC connections transfer power, but transfer ZERO SRI to other grids.
.
Reactive Power
No AC grid can function without positive reactive power; say power factor of 0.8
Wind/solar systems take reactive power FROM the grid; say power factor of -0.8
All traditional power plants are automatically set up to provide positive reactive power TO the grid.
.
Synchronous Condenser Systems
The weather-dependent, variable/intermittent, wind/solar feed-ins to the grid often create transmission faults.
Those faults are often minimized with synchronous condenser systems that provide positive reactive power TO the grid.
Blackouts
In case of too much W/S power, it needs to be curtailed.
Owners usually get paid for what they could have produced.
.
In case of too little W/S power or a W/S outage, reliable, quick-reacting CCGT plants, in Hot Synchronous Standby, HSS, mode, would provide:
.
1)) Instant SRI to the grid for “ride-through” to give switches time to switch, and
2) Provide power to the grid, within seconds, to counteract voltage/frequency drops due to W/S outages, 24/7/365; if battery systems were used, they would be empty after a few hours, with no prospect of a black grid to refill them.
.
Spain/Portugal would have needed about 10,000 MW of CCGT plants in HSS mode to avoid its recent blackout.
They would operate at 50% output throughout the year, and quickly provide up to 5000 MW, in case of a W/S outage.
Of course,no subsidy, no w/s! It was always so. Now, 15 $Trillion later, we have wasted our seed corn.
More problems=more government. Rinse and repeat.
It’s not quite that simple, Nick. Batteries have limited charge and discharge rates, and inverters have limited power and current ratings, as well as imperfect inertia simulation algorithms.
Inverters carry the whole of the power from wind and solar, and also DC interconnectors etc. Batteries can deliver a lot of power. Imperfect algorihtms can be got right.
And then there was the Iberian blackout…
France’s grid has a lot of inertia and that saved them, enabling the power to be restored relatively quickly.
Not so in Espana and Portugal.
By your metric, a green success.
Yep. But you forgot to mention that under your plan, all you have to do to keep it running is to print dollars and hand them out.
And once again, Nick leaps for the simple solution without ever trying to actually understand the problem.
Have you ever thought to think about why they haven’t already implemented these solutions, since to your mind, they are so simple?
and are really expensive to solve a problem that doesn’t need to be solved with scam energy
None of these comments deal with out-of-step, i.e., phase problems. You can have a generator operating at the correct frequency and voltage yet be terribly out of sync with the grid.
In todays world, that is dominated by synchronous generators with massive inertia this is a problem when bringing a generator online. There are controls and procedures that control rooms have to follow when doing so. Otherwise, you could connect a generator 180 degrees out of phase and cause massive problems. This is with a small number of large generators.
Imagine having 100 times the number of grid forming inverters connected to the grid. How are these synchronized as some number are added back onto the grid after a large fault? My research into grid forming nverters leads me to believe that they operate asynchronously at best. How is a control hierarchy established? Is one grid forming inverter designated as the master that all others follow? How is that communication accomplished. I foresee massive cascading failures as multiple grid forming inverters are reconnected to the grid with no process for controlling phase differences.
No you cant. In the case of rotating masses, they literally tear themselves apart. Bringing a generator online is a very precise process where the frequency and phase is very precisely matched.
You could try to force it with an inverter, but why would you? The whole point of grid forming inverters is to stabilise the grid, not destabilise it.
I think you’re missing the drift of what Jim was saying. He wasn’t saying it’s okay that the generator is out of sync; quite the contrary. He was saying sync is still an issue even if frequency and voltage are correct.
Exactly! People don’t realize what happens when one generator is crossing zero while another is at peak voltage even if the frequencies are EXACTLY the same.
The grid isn’t a wild west of voltages.
Sure it is. Voltage, current, reactance, phase, faults, failures, etc.
Have you ever been in a power plant control room? How about a distribution control center?
Here is just one article about control centers.
https://www.tdworld.com/smart-utility/article/21265807/the-control-room-is-at-the-heart-of-the-energy-transition
It was in Spain. Lots of reports of sobretension (over voltage) out of scope. Almost certainly the causes of the initial trips.
Do you have a reference? I’ve not seen anything definitive about the causes yet. The interconnectors dropping out isn’t a “cause” on its own. Which probably means they either cant locate the source of the issue/fault or for whatever reason cant disclose details.
Start e.g. here:
This has the ring of truth about it for the Spanish Apagon:
Desde el pasado mes de enero, tal y como adelantó elEconomista.es, Red Eléctrica afronta un problema en el control de tensión de sus líneas. La creciente integración de energías renovables y la caída de demanda provocan unas elevadas oscilaciones en los niveles de tensión que podrían llegar a estar detrás de la causa de este apagón.
https://www.eleconomista.es/energia/noticias/13337393/04/25/espana-sufre-el-peor-apagon-de-su-historia.html
Since January, Spanish grid operator Red Eléctrica has been facing problems with voltage control on its transmission lines. The growing production and integration of renewable energy and falling demand on the grid (due to domestic solar) have caused excessive voltage oscillations which could have caused the blackout.
Formal work is being heavily impeded by political interference. Even ENTSO-E is not able to access key information. That is why early whistleblower reports in Spanish probably offer the best guide to date. Online translators tend to make a hash of technical Spanish which led to misreporting in the English speaking world.
There is also a report of frequency traces leading up to the event from various locations around Europe including Malaga: these show the effects of inverters fighting each other on different sync. The consequences of out of sync resonances is amplified voltages.
https://gridradar.net/en/blog/post/blackout-iberian-peninsula
Isn’t a cause.
And neither is this.
A cause is a specific instance of a problem that caused instability that lead to a cascading failure. What you have posted is speculation.
Look at his comment below. He’s just said two generators could be on the grid out of phase. Way out of phase in his example. It doesn’t happen.
You don’t read well or you enjoy misquoting. I asked what would happen if denerating plants simply connected to the grid without synchronizing while offline.
How do you synchronize thousands of grid forming inverters that no longer grid follow?
It is a simple question, why don’t you answer it?
Your question doesn’t even make sense. They reinforce the grid signal that is present. One has to go first obviously.
Out of 100 grid forming inverters, which one goes first? Again, I am reminded of a NASCAR race to the finish line!
Is this really your main concern? The people who run the grids have protocols and procedures. Starting a grid is non trivial but it’s not because they’re playing whackamole with rogue generators.
Exactly, that is why they drop off line. It takes a serious fault for this to occur with large inertia rotating masses.
Here is a good document that describes the need for frequency, voltage, and phase synchronization before connecting to the grid. This is normally done offline so the moment the connection is made there are no shocks.
https://www.testandmeasurementtips.com/how-ac-power-sources-get-synchronized-faq/
Now explain how 100’s or 1000’s of grid forming inverters are synchronized as they are recovering from a grid shutdown.
It’s almost like you think frequency control is easy with rotating masses. It’s not and we’ve had many, many grid outages prior to renewables.
Once a stable frequency is established, they’re good to go. The Spanish grid was back up in 10 hours which is unexpectedly great recovery. Many inertia heavy grids have taken longer.
A stable frequency has nothing to do with synchronization of phase!
Grid forming inverters don’t use phase locked circuits otherwise they would be grid following inverters. Explain how synchronization is accomplished.
It’s an algorithm. If frequency of the grid is decreasing then I expect grid following inverters will lead not follow the decrease. I can’t give you specifics but suffice to say it works even if not optimally yet. Getting the right response with the right amount of damping is key.
I meant of course …
If frequency of the grid is decreasing then I expect grid forming inverters will lead not follow the decrease.
Grid following inverters will follow not lead.
They can only lead if they have the overhead capacity to add to the grid. Over the length of a few cycles this may be possible. If however, something like a solar farm goes out due to a hail storm, can they replace the capacity for hours or days?
That is why many fossil fuel plants typically run at something less that 100% capacity. That provides the overhead for long term problems and maintenance outages.
Any generator is in this position. A battery can increase its output essentially instantaneously. Obviously it can only supply its maximum output. That’s a trivially obvious statement.
The Spanish grid doesn’t have grid forming inverters, and wind and solar were kept off the grid while it was energised by French nuclear, CCGT and hydro. It also lacks grid batteries, relying on pumped storage instead (which also provides inertia even when pumping).
Renewable grids around the world haven’t focused on stability. They’ve focused on adding renewable energy generation. I expect Spain was a wakeup call.
The evidence suggests that grid operating standards in grids where renewables are supposed to be prioritised are happily compromised to accommodate renewables. That’s the case for ENTSO-E which has a 2 second inertia standard which wasn’t enough to keep Spain on board, despite their claim that they were running at 2.3 seconds at the time. We see the same thing in the UK where the smaller demand levels mean you need a higher level of inertia in seconds to cope with say the loss of an interconnector.
In the context of this article, all that means is you load shed maybe a second later if the remaining generation can’t support the load. And if it can then the reason inertia saved the grid is because there was insufficient fast response generation.
Sorry Nick, but you’re just plain wrong on this. Grid inertia is the maintenance of a store of kinetic energy in order to maintain a consistent frequency on the grid. Inverters can’t provide that, nor can all the algorithms in the world simulate it when there’s no external signal to level against. In the absence of spinning inertia, inverters can’t provide that signal to one another because they are all attempting to synchronise against an underlying signal that doesn’t exist. Absent that single source of truth, they will constantly chase synchronism with one another in a positive feedback loop that no amount of algorithmic magic can prevent. It’s fundamentally an expression of the halting problem.
And no, a flywheel can’t provide that source of truth. They are net energy consumers and require an external energy source to maintain their spin. They can provide a small amount of energy over a short period (far shorter than it takes for them to spin down), but at a non-synchronous frequency, which is useless for frequency maintenance and consequently useless for supplying grid inertia.
Yes they can, they’re called grid forming inverters and that’s precisely what they do. Most inverters (eg rooftop solar inverters) dont but that’s by design choice.
They “mimic” synchronous inertia, but they cannot and will not replace it. The spanish grid was taken down, at least in part, by grid-forming inverters chasing their own tails and inducing rapid oscillations to the grid frequency.
No it wasn’t. Do you have a reference to back your claim?
Different sets of inverters were seeing a different phase and ended up fighting each other. There are no grid forming inverters in Spain AFAIK, just as they don’t have grid batteries. – just pumped storage and hydro along with nuclear, CCGT, solar and wind and a token small coal station.
I think you have misunderstood.
To hold a frequency up under overload takes energy
If the renewables are not supplying that energy no inverter can fix the problem.
Only battery powered inverters can, at huge expense
And for a good few minutes too. Lithium power packs can survive near total discharge in a few minutes so peak power available is quite large.
Batteries do solve a large part of the spinning mass problem, but the cost is horrendous
Yes. And prices continue to drop.
That entirely depends on their capacity compared to the load they have to carry. If it’s for the purposes of grid stability then it doesn’t take much.
Batteries cannot do so reliably.
Too many factors mean batteries, like solar voltaic and wind turbine generators, do not deliver their name plate capacity.
In addition, there is the voltage turn-on phenomenon.
I will not bother to go into phase lock loops that allegedly make inverters grid forming. Nor will I repeat prior posts about the risk of depending on internet time or GPS time to synchronize systems.
Grid following inverters use phase locked loops.
Grid forming inverters use may use a PLL for monitoring but use some sort of algorithm to define their output because they can lead the voltage to simulate inertia.
And therein lies the problem. Out of “n” grid forming inverters, which one is the boss? My guess is they all compete against each other.
Why? Part of their algorithm is to maintain frequency and they react if that changes. If whatever signal is seen is slowing then lead. If its speeding up then lag. Do it in such a way as to dampen oscillations.
Now why do the “n” grid forming inverters compete? Spell out where you think the problem comes from.
If the circuitry has access to an accurate time signal, like one of the satellite positioning systems, (GPS , Glonass, Beidou, Galileoj, then synchronisation of generated power is no problem.
If there is voltage on the grid, then the inverters lock on to the signal.
There are increasing amounts of electricity being transferred by HVDC (high voltage direct current). There are advantages and disadvantages, but one of the advantages is that connecting asynchronous grids or generators is easier, as the synchronisation is to an external source, and the DC is converted to AC locally.
A 50 Hz AC signal has a wavelength of 6000 km. The electricity on the grid is travelling at nearly the speed of light, In a grid spanning long distances – say 6000 km, the phase of incoming electricity may be unsynchronised with local generation, even though both sources a perfectly in synch with each other.
So it’s not a simple matter.
Provided the available power is in excess of the load, then any problems are the usual ones – equipment failure, accidental line shorts or opens, chaos, and so on.
Unfortunately, economics and cost effectiveness enter the fray. Renewables are no problem, provided the rest of the grid can meet peak load by itself. In that case, what use are renewables? A very expensive waste of time, effort, and money, if the grid has to provide enough power to do without them.
Many questions, few answers. I have great respect for the people who manage to provide reliable, affordable electric power to me. Next to a reliable and sufficient supply of potable water, electric power is the basis of civilisation.
Put the money wasted on “climate change” into ensuring reliable and affordable water and electricity supplies. Who could possibly object?
The satellite time signal is a nice idea but it doesn’t work – the absolute phase angle on any part of the grid can differ because of impedance and phase angle change through intermediary transformers.
Your local grid (if you are a generator) has to be up and running before you can synchronize to it and jump on.
So you are dependent on starter thermal, hydro or nuclear plants – in the Iberian peninsula they had the French interconnectors as a base line.
I’m pretty sure you would want renewables to stay out of the way until the grid was re-established.
I don’t think a “black start” from renewables is remotely possible.
Not directly, I agree. But a black start from battery backed, grid forming inverters would be possible.
At any moment, this is true but at all times the generators are trying to maintain the exact frequency. There is no reason they couldn’t be trying to achieve the exact frequency in line with an independent time signal.
You have to look at the phase at the edge of the network as you expand it. You can’t just sync to a time standard. Moreover inrush current is going to alter the frequency when you throw the switch.
Why not? The grid may move away from it but can always move back towards it as it settles just as the grid always strives to maintain its frequency very precisely.
Using a common time signal would simplify joining islanded grids and AC interconnectors because they’d already be in phase and would only need to wait for the moment when they’re precisely in phase to connect.
I dont know whether the advantages would outweigh the added complexity in practice but it certainly seems to have advantages.
Because the phase is different at the edge of the grid. Chasmsteed has it right.
And why does that matter at the generation end where the time signal could be used?
Because things like transmission and distribution lines, substations, industrial sites, loads of all kinds can move the phase around as you move around the grid. 100 mi down the road can be different than you.
That is why most large fossil fuel power plants still maintain the old 3 lamp unit that shows synchronization is correct as a backup to present controls. Matching is done while the generator is offline while making the match. Then the generator is connected.
And? This doesn’t even begin to answer why the generation end can’t maintain a standard time signal. If two generators necessarily have different phases on their specific network then obviously one would run to an offset.
This is not an appropriate forum for teaching power plant operation. Here is a link to a good document that explains operational problems when out of phase generators are connected to the grid.
Be sure to read carefully all the catastrophic things that ca happen.
https://electrical-engineering-portal.com/preparing-to-synchronize-a-generator-to-the-grid
Thats ok Jim, we’ve discussed this before
Grid forming inverters. Sounds nice, but that is not practical.
There is a problem in Europe today as someone has disrupted GPS over large areas. Clouds also affect GPS reliability.
Yeah Nick.
Where’s Rube Goldberg when we need him most?
We understand Nick. You’re entitled to your religious beliefs. Faith, flywheels, and batteries abide, but the greatest of these are batteries.
Also during the discussion of cascading failure she never mentioned load shedding. Instead she said it wasn’t what happened in her example because the “UK’s inertial level was higher” and that’s just rubbish. High inertia levels doesn’t “stop” cascading failures. Load shedding does that.
On 22nd December 2023 the UK had a cascading trip that started with the IFA1 interconnector and then took out Cottam Power Station, and echoed all the way up to Northern Scotland knocking out the Caithness HVDC link and a bunch of wind farms. Total acknowledged loss of 1638MW. The nadir was 49.273 Hz, somewhat above the to LFDD frequency of 48.8Hz (which was reached in August 2019, resulting in load shed that actually shed addional renewabkes generation as well: the overall loss was up to 3GW). Kathryn is correct.
How is Kathryn correct when she didn’t even mention load shedding? You’re going to have to explain to me how a cascading failure that is arrested by load shedding is better for having more inertia.
If the energy isn’t there, it isn’t there and that’s why failures cascade and it gets worse with each failure.
If you have more inertia you can avoid getting to load shedding at all. In Spain the load shedding didn’t help. The first load to be shed is pumped storage pumping which was actually contributing a significant amount of inertia to the grid, the loss of which speeded up the frequency decline. The increase in the speed of decline will have helped trip more generators offline. The result was like a hot knife through butter: LFDD was outrun by generator disconnections.
Not in the context of a conversation about cascading failures. If you’re in a cascading failure situation then inertia isn’t stopping it.
Kathryn’s discussion didn’t even mention load shedding which is the fundamental solution to cascading failure. That’s a massive misunderstanding on her part.
If this happened, the load of the storage pump wasn’t even close to being enough.
And this has happened many times in the past. Inertia didn’t save pre-renewable grids in the past either.
Yes, obviously inertia stabilises grids and prevents them from failing but this is a response to the specific explanation Kathryn made about cascading failure.
Synchronous Rotational Inertia, SRI, provides a few seconds of “ride through”, which gives switches time to switch, and for SRI hot standby power plants to instantly feed electricity into the grid, to reduce/eliminate voltage and frequency drops, and for automatic load shedding to take place, as needed, etc.
Spain’s problem was not enough SRI and not enough SRI hot standby capacity, MW, and not enough automatic load shedding.
The result BOOM
The Spanish grid and generation mix will require expensive upgrades/changes to ensure the standard 99.98% reliable delivery of electricity, unless Spain—and its neighbors—are comfortable with increased frequency and severity of brownouts/blackouts.
Black Start Procedure for a 100 MW CCGT Power Plant
Initial Power Source: The on-site auxiliary generator is started. It provides power to critical plant systems, including control, safety, and communication systems.
Plant Startup: The auxiliary generator then powers the CCGT plant’s essential systems. This includes cooling systems, fuel handling systems, and starting the gas turbine.
Connecting to the Grid: After the CCGT plant is spinning at 3600 rpm at the same phase and frequency as the grid, it can be connected to the grid to supply power to its section of the grid. That section powers another power plant, etc., until all sections are up and running. Only then, grid-destabilizing W/S systems are connected.
How many milliseconds in “instantly”?
Wind and solar take positive reactive power from AC grids, which means they destabilize AC grids.
SRI traditional plants are AUTOMATICALLY set up to feed positive reactive power TO the grid, thereby stabilizing AC grids.
Some haven’t got the memo…
Green inertia’ projects and world-first tech tell British energy success story
https://www.neso.energy/document/246426/download
Flywheels….
Yes, if you want to stabilise with flywheels, they don’t have to be attached to a steam engine or some such. They are just an energy store which can be topped up from any power source, and will be backed with a battery, which gives much better control.
It’s a completely stupid strategy – I’m sure, at least I hope, you realise that.
You don’t seem to have anything useful to say about it.
You don’t find reality useful? Really?
No wonder you believe in the unbelievable: Unreliables.
Please be careful strat, there’s no longer free speech for blasphemy in your country (unless it’s blasphemy against Christian faith).
hmmm… adding a real flywheel to wind or solar farm seems like adding another layer of epicycle to a fundamentally unsound system
+10
Nick,
A flywheel can do one of two main functions.
IF it is to be used as a short term energy store then it will NOT be synchronous. It will be no different to a battery. It will need an inverter to convert to grid frequency.
IF it is synchronous, spinning at 3000 RPM for a 50Hz grid, then it has a very LIMITED amount of energy that it can provide until it too slows down to BELOW the allowable grid frequency range. Obviously, the heavier, the better.
Inertia was not really covered in the video. It could have been done a lot better. Inertia is the instantaneous ability to absorb or delivery energy to maintain the desired frequency. Just one simple example would have made a huge difference. eg a spinning 1000T mass, (distribute evenly with a mean diameter of say 3m), calculate the energy required to spin it from 3000Hz by +/- 15 Hz, (0.5% variance). That is inertia. The resistance of that mass to change speed, involve large energy transfer(s).
Have a play here and look at the incredible energy stored for a typical turbine generator.
http://www.calculatoredge.com/mech/flywheel.htm
Run the numbers, calculate the energy difference for the same mass at 3000Hz and 3015Hz. Now that’s inertia. No battery powered inverter is going to step up to that delivery, (over milliseconds).
A battery inverter can do that.
At horrendous cost.
If you want inertia from a battery/inverter system, then use them to spin up a large spinning mass, (using DC motors), and take the AC from a connected generator. Anything else is doomed to fail due to the massive currents involved.
Just look at the current required to supply inertia when using a battery to deliver massive amounts of MJ in the milliseconds timespan.
As an example, just ONE megaJoule in 1 milliseconds would require 1×10^12 watts. If your battery was 1000V, then you only need to carry 1×10^9 Amps. ONLY.
Let me know when you find a conductor to carry that sort of power. And of course, the battery that feeds it. Using a very large stack of parallel batteries and inverters works on paper BUT then you have to ensure that they all work perfectly in sync. And don’t themselves make an oscillator between themselves. Which they will because even a signal can only propagate so fast down the wires.
+10
No. Batteries are electro chemical and have chemical reaction times.
The voltage turn on delay for an cold or old battery would fail to deliver.
You do realize that whatever power source keeps the flywheel turning must supply sufficient energy (plus losses) to provide the output power that is needed for whatever time is necessary, right?
Does that sound like a typical fossil fuel power plant to you?
As an engineer with an EE/ME degree but zero experience in delivering electricity to consumers, my analysis of electricity supply grids is quite simple: They can’t possibly work. Grid operators have no control whatsoever over demand, and scant control over supply in a system where the laws of physics dictate that the input precisely equals the output when line losses are included. Yet, they must supply power with extremely tight control of voltage, frequency, and even number of revolutions of the supply generators (or equivalent solar and wind cycles) to about one revolution per day. I defy anyone to describe a control system that can do that.
“They can’t possibly work”
They have been working reliably for a long time.
As one who spent 10 days in the Texas grid shutdown a couple of years back, I dispute that statement.
They have been working reliably for a long time.
Even despite Dunkelflaute?
Cold without wind: German ‘dunkelflaute’ brings electricity prices to crisis levels and depletes gas reserves
https://english.elpais.com/international/2024-12-12/cold-without-wind-german-dunkelflaute-brings-electricity-prices-to-crisis-levels-and-depletes-gas-reserves.html
Why did they mention gas, Nick? Your unreliables should have been able to cope…
I have no idea about what you are talking.
You heard about redispatching costs soaring?
That’s what you call reliable?
You live in a different world.
Have you ever been in the control room of a really large fossil fuel power plant? Have you reviewed the training documents and procedure information for maintaining proper operation?
Are we going to have similar control rooms for all the grid forming inverters that no longer follow the grid with phase locked loops? I envision a NASCAR race at Daytona where everyone is trying to lead. Not a good look for the grid operation.
Where TJ’s analysis of grids got wrong is ignoring the role of inductance in the operation of generators and transmission lines. This makes an AC grid a whole different beast from a DC grid. OTOH, most people who haven’t had instruction or read up on electric power systems typically know Jack Scheiss about grid operation.
In a DC grid, power flow is determined by potential differences between the various nodes in the grid. The equivalent of inertia can be supplied by capacitors that absorb power when the node voltage is rising and supply power when voltage drops.
In an AC grid REAL power flow is largely determined by the PHASE difference between nodes on a transmission line and REACTIVE power flow is largely determined by VOLTAGE differences (but note transmission lines typically absorb reactive power). The power output of a generator also varies with the phase difference between the unloaded phase (shaft angle) and the terminals connecting to the grid.
The control scheme for a synchronous generator is relatively simple, the excitation is adjusted to maintain terminal voltage (generator inductance limiting current) and the power from the prime mover is adjusted to maintain frequency, though with a drooping response to promote load sharing between generators.
Inertia comes into play when a change in load occurs. An increase in load will use the inertia to supply the extra power with the consequence of reducing frequency, which then results in a change in relative phase with other nodes which then results in more power flowing into the node (load sharing). The governors on the prime movers will then respond by increasing mechanical power (torque) to the generator shaft.
The inherent problem with most inverters is that they are effectively voltage sources, and the control algorithm for controlling reactive power is more complicated and needs to respond much faster than voltage control on a synchronous generator. A synchronous generator will survive having its terminals shorted for a few cycles with no changes in excitation, but the same is not true for an inverter.
The inherent problem with most wind and solar operations is that they usually do not have any significant local energy storage to respond to changes in frequency and this is compounded by many if not most such installations requiring the grid to respond to changes in plant output due to changes in wind or light reaching the solar panels.
And the solution is grid storage in some form. Some locations can have pumped hydro but anywhere can have batteries. It’s part of the renewable solution. It’s not unexpected or optional and it is costly.
Fossil fuels have been very effective and cheap in the past but that’s going to continue to change as our global demand for energy increases and so do their production costs and scarcity.
Nuclear is even better, energy storage would eliminate much of the need for daily load following.
Battery storage still needs to prove that large scale battery fires are a thing of the past.
Lithium probably isn’t the best solution for grid scale batteries. Something cheaper with better longevity at the cost of increased weight would be better. It doesn’t matter if the energy density is low compared to Lithium if the battery is stationary. Its even better if it uses less flammable chemistry. I’m still betting Sodium based will win eventually.
Mr. ToolMan: So many comments from you, why don’t you just cut to the chase and say you believe that wind and solar will power the grid AFTER the miracle enters your formula?
I dont want wind and solar to “win”. I want fusion to win. But reality is that the problem of fusion hasn’t been solved and is proving incredibly difficult.
If nuclear energy increases, then great. But its not yet.
It has nothing to do with CO2 levels and everything to do with fossil fuels eventually running out and us needing to do something about it before that happens.
If we wait until fossil fuelled energy production cant keep up with demand and allow market forces to force the change, then many of the earth’s population will die because by definition there wont be enough energy to sustain everyone.
Point is valid, but we do not need an emergency transformation as late as yesterday. Human ingenuity, technology advances, new discoveries are the path of the future. Eventually, coal, gas, and oil, even fission, will be replaced with something better.
It is a shame we cannot be allowed to hope for the future.
While we’re doing it to “address the climate emergency” we’re doing it for the wrong reasons and that impacts the pace.
Having said that, we’ve spent 20 years and barely scratched the surface. in terms of actual transition Much of what has happened can be categorised as “learnings” and they’re necessary time and money being spent.
There is no “something better” solution without actively working towards it unless we’re very lucky and someone unexpectedly stumbles onto something entirely new.
Your admission of utter ignorance of how the grid works is manifest in your misguided statement that it can’t possibly work. That video is missing a host of critical detail and glosses over relevant engineering facts. Not only do rotating generators possess inertia, but the power sources driving them are throttleable to various extents, rather rapidly in some cases and much slower in other cases.
So it is a mixture of humongous inertia of rotating masses, plus the turbines driving these rotating masses can be throttled. That is in essence how one can provide a control system to match electricity output to demand.
To simplify an analogy, if your ICE car is traveling 60 mph on level ground and encounters an uphill slope, you simply add more engine throttle to maintain speed. Or of you encounter a downhill slope, you reduce throttle to maintain speed.
Now there is a reason power grids are generally composed of multiple different prime energy sources, because some of them can throttle very quickly and others need considerably time to throttle up or down. You need the fast ones for rapid load changes and you need the slow ones for constant conditions. And I might add solar and wind sources do not have the ability to meet these inertial and throttling responses.
The primary reason frequency must be within a few percent of mains standard (60 Hz in N America and stupid 50 Hz everywhere else) lies in resonant behavior of the huge electromagnetic generators, which the big ones do NOT turn at 3,000 rpm as the micky mouse video above suggests. (only small capacity gensets operate at 3000-3600 rpm, the big mains gensets are typically 20-30 feet in diameter and operate at much lower speeds, the speed being determined by how many magnetic poles there are and limited by the tip speeds of the large diameter masses)
The big ones, encounter destructive resonance when the frequency falls to a few percent below mains and will literally fling themselves to bits. So if the speed of a big genset falls too much, brakes are applied and it trips off, to prevent their self destruction. This resonance is a combination of mechanical effects and magnetic ones as there are fore and counter (attractive and repulsive) forces as each magnetic pole passes through the changing fields.
Back to the throttling aspect, the fastest to change shaft input power are hydroelectric generators, by opening and closing the gates in the water turbines. Well if you had smaller diesel gensets they too can respond rapidly to changing loads…. Next are gas turbine gensets, where just as with airliner gas turbines, you can throttle them, from about 65% power (below which a gas turbine will not stay running) to 105% for short periods. Then we have the various boiler operated steam turbines. Oil fired or Gas fired can throttle, but have longer response times, although there are steam reservoirs to provide some throttle buffers – recall steam locomotives can throttle though the fireman must stay on top of adding fuel to match demand.
The worst response times are with nuclear reactors heating water for driving steam turbines. they can be throttled as well but not nearly as responsive or with as wide a range as fossil fueled boiler/steam turbine plants.
To give some perspective on the response times, if there is a cascade tripping off of mains gensets then the time to restore mains for various plant types is as follows:
Hydro – 20 minutes
Gas turbines – also 20 minutes
Gas or Oil fired boilers and steam turbines – 8 hours if the boilers were shut down.
Nuclear plants being scrambled off – 2 to 3 days
This does not mean these have throttling responses that slow, they can throttle up and down rapidly, but if they are shut down and have to do a cold restart, the times above indicate their response times and in a sense their thermal inertia.
So how can mains systems provide control systems as power generated must match the load? Because of the rotating inertia and those prime movers can all be throttled.
Next are gas turbine gensets, where just as with airliner gas turbines, you can throttle them, from about 65% power (below which a gas turbine will not stay running) to 105% for short periods
Gas turbines are stable to 50% load
Below that, they are unstable
For ramping service, aka babysitting for wind and solar, they are operated at 75% to ramp up 25% and ramp down 25%, as needed
Depends a lot on the turbine and whether it has a steam circuit as well, but the broad brush is about correct.
Hydro comes on way faster though. Worst case 10 minutes from cold. If they are already spinning and synched up a -a lot faster than that.
Coal and nuclear are the slowest.
Dinorwig gets to 1.7GW from spinning in air in 12 seconds.
According to the Wiki,
Not bad. There are 6 of them.
And then the Wiki quotes 16 seconds if they’re already synchronised
And the reference they used quotes “First Hydro Companies” page
So a bit less than the 1.7GW in 12 seconds but still impressive.
Not bad.
Maybe 3-10 secs under light overload? Less than 1 second under heavy load?
Quickly disconnecting is your only option, and that’s what the electronic protection will do – within milliseconds of detecting the need.
And disconnecting megawatts quickly is not easy – the bloody juice wants to keep flowing. It has inertia all of its own! You can’t win, sometimes.
I really thought he was trying to be funny, turns out I got the joke (grid can’t work but does!) he didn’t intend.
Synchronous power plants provide enough inertia to make the system work, already for 120 years.
It is complicated to non STEM people, who unfortunately have seized power, because they know snake talk.
Adding grid-disturbing wind and solar requires more and more “measures” to make the system provide electricity service to a modern society with a standard reliability of 99.98%
There has been much talk about frequency. This is the symptom of mismatch between supply and demand in a system which contains stored energy in the form of rotational inertia. In principle it can also be provided by battery/electronic systems but there are significant limitations. Using GNSS timing in place of grid synchronism is a possibility.
However my comment is to repudiate the statement that the generators have no control of either supply or demand. The latter is increasingly part of the whole demand managed sector, EV, Smart meters, regulated loads. In addition there is an alarmingly simple way of reducing demand – reduce the grid voltage. All devices with power factors around unity will reduce their energy needs if the voltage is reduced.
Finally, it has become apparent that the electrical systems in developed countries serve two types of loads, informational and thermal. Apart from the stupidly named AI systems which spend vast amounts of informational power extracting garbage from garbage, most information systems have been designed to use less and less power. Also, it generally requires a DC supply so perhaps there is a case for consumers to have two meters. We will not die of cold or over temperature but losing your internet or mobile might be life threatening.Deliveroo can then replace your non-functioning electric oven or microwave. Oh, sorry, their EV vehicles have just been switched off.
Summary, the whole renewable Net Zero is doomed.
The have humans monitoring 24/7/365 and making adjustments in real time.
Video clips are a superficial, poor means of education.
The written word, with appropriate equations, takes far less bandwidth for far more information.
It is only by slathering corrective item-on-item, none of which are as good as the original system, that the intermittent renewable energy (IRE) system is supposed to become as stable as the system which has worked and still works. France shows that clearly – nuclear Rankine cycle plants = spinning inertia. Even France has caved, however, turning off Phenix after it became successful and now spending $BILLIONS to dismantle it.
IRE systems require orders of magnitude more computers, more electronic grid-forming inverters – widgets, and non-power producing rotational inertia. These are ‘lipstick on the IRE pig’.
There is NO bacon on that pig. It is pure graft.
A glimmer of hope for the future:
Old style alarm clocks store energy of motion, after you wind them up. It’s not inertia. But no I’m not proposing to plant a bazillion alarm clocks around the countryside.
Thanks, that was helpful.
As an engineer who has worked in that industry I give the presentation a B-. You will leave knowing more than you came with, but a few of the details would be wrong. I think it was a matter of not explaining how Cacao is harvested to explain how to bake a chocolate chip cookie. There is too much to explain before explaining can start.
Here’s how NESO are playing fast and loose with inertia in the GB grid: points that lie below the red line are in breach of their inertia commitment. They are asking for formal permission to reduce the inertia to the orange line, which they have already been illicitly experimenting with. It is not hard to imagine that they will in turn breach that if the bonus for a record low fossil fuel generation statistic is high enough. Meanwhile we pray that nothing goes ping while they are playing.
At least its happening at low demand levels.
Accurate. It’ll be a failure that brings the whole thing undone, and unlikely to be during steady-ish state supply.
Low demand levels are the riskiest because any trip represents a bigger share of the load – like Friday’s 1.4GW trip of the NSL interconnector. Fortunately wind was not strong so there was enough CCGT to limit the frequency decline and more added to replace the lost supply, but with batteries being also an important line of defence.
But conversely a fast reacting gas turbine can ramp up more of that load more quickly. But yes, large changes, even if only relatively, are problematic.
Unless we change direction, there will come a point where there are sufficient batteries stabilising the grid that we’ll be better off than when the grid was inertia heavy.