Induced atmospheric oscillations.

Firefox (with UBlock origin) -> Private Window -> Reader Mode works for me.

I then Select All -> Copy -> Paste into my favourite document editor (Libre Write) and save for future reference.

Article contains too many technical phrases for greenwasher to comprehend eg frequency

Perhaps, you could explain what inertia means in this context. All I can find is that turbine generators have angular momentum and hence inertia. This could have two implications. Firstly, changes in power are gradual. They don't just switch off like a light switch. Secondly, it could mean they act like a battery, you can trade some of the angular momentum for more power, almost instantaneously. I'm not sure which people are referring to.

Now, it seems to me that lithium batteries are good for more power delivered in an instant. It seems obvious, a modern grid should be provisioned with enough batteries to achieve this. It is a uk.d-i-y poster favourite to point out batteries are unsuitable for long term storage, but they should be suitable for smoothing short term fluctuation.

So potentially this sounds like poor grid management, rather than an inherent problem with solar.

FWIW. I remember my Dad talking about the goal of nuclear electrical generation without steam turbines, I'm not sure now what he meant, as he died when I was in my mid-teens. So it isn't necessarily just renewables that introduce problems. We shouldn't be Luddites, just because you don't like the dislike the renewable political bullshit.

You don’t seem to realise that the problems lie with the (inertia-less) inverters. Battery storage requires inverters.

Yes, I don't understand. Which is why I'm asking. Do you understand, could you explain.

What is the problem with inverters, explicitly?

The trouble with panics, and journalists, is that people pick up on buzzwords and bandy them around, without understanding what they mean.

But all that inertia does is to allow a small amount of back-driving using the kinetic energy in the rotating parts? Why can't an inverter do just the same if it has a local energy store? Isn't is just a matter of an inverter with a small amount of available battery storage and appropriate software? It should be easy to emulate what rotating hardware does.

nib

From the pdf Clive mentioned, they trade angular momentum for maybe 10s.

That is one alternative mentioned, with no examples of use; maybe Musky's

Spain seems to have solar during the day, wind overnight; hydro and pump storage available, but not for the first 5 seconds.

The problem with inverters is that they have no inertia.

If there is a sudden fluctuation in the grid frequency that goes outside the set operating range, the inverter will protect itself and the grid from what it sees as a fault condition by disconnecting. This makes the problem worse because as the energy supply is reduced, the grid frequency will fall too, tripping other inverters. Under these conditions a battery storage facility won’t even start up. This type of grid is inherently unstable.

But the spinning turbines of the <normal> generators have by comparison a huge amount of inertia; if the grid frequency drops on the timescales that trip inverters, the spinning turbines will just carry on as if nothing has happened, putting energy into the grid at an unchanged frequency. This kind of grid is inherently stable.

The Spanish were running on 78% renewables, when an unspecified event started the inverters tripping out, so that suggests an upper limit for the proportion of renewables on a grid.

C’est la vie, say the old folks.

AIUI that battery is doing a bit of load balancing. It's the cost of that sort of battery, scaled up, that gave me the trillion quid figure.

It's a more general inertia not of particular mechanical parts, but of continuity of supply. In general, apart from actual breakdown, non-renewable generators run continuously and can be expected to supply electricity indefinitely (in the short term). Shutdown for maintenance will be scheduled, and grid operators can ensure continuity of supply.

Renewables don't offer continuity of supply. A renewable power source can lose half its output in a matter of minutes, and in an unplanned way. OK, we know there's going to be no solar at night, but during the day, thick clouds can gather over a large area. Wind can rise and fall in a fairly short time, particularly around heavy rain. That's the kind of lack of inertia being talked about. It's not really the inverters as such.

Gathering renewable energy over a large area will mitigate this to some extent, but I don't think the European grid (or the UK one) has sufficient capacity, enough redundant links, to transfer power in varying directions over large areas very quickly. Many more cables everywhere will help here, although the fundamental problem is that the good places for gathering renewable energy are by definition not the places where most power is used. This is particularly true of offshore generators.

Yes, I don't understand either. With rotational generators if the load increases beyond what the generator can supply then it rotates more slowly, i.e. the frequency drops. The managers can monitor this and try to provide more generational capacity or in emergency shed load.

But with DC interconnectors, wind, solar cells, or batteries, all of which need an inverter, there is no need for the frequency to drop and indeed the electronics should keep it stable. If the load is too great then the voltage will drop. That's exactly what happens to the solar cells on my roof, and when it drops too much they stop feeding my solar power back to the grid. So on a national level if that happens then the system might produce "brown outs" but I don't see any reason for grid instability or indeed frequency changes. So I'd appreciate an explanation from an expert.

You may think it's funny, but he has a point.

I didn't realize the planning for ours, had started so long ago. It could be ten years, until the projected completion date of 2028. No ground has been broken. No announcements made. To meet a 2028 date, while the current date is "mid 2025", you would think there would be some sort of visible indication of activity. If it take five years here, to build a public beach house, they are not going to hurry to pour a concrete base for an SMR.

It's nuclear, and it comes with its own bureaucratic hive.

For one of the recent reactors, the paper associated with the reactor design, was something like 2 million sheets of paper. Maybe an AI could read that and make a three line summary ? Hoping...

Paul

That's normal. Nothing wrong with that. Take all the time you need, to do it right, not do it twice.

It's engineering. And the purpose of working it out, is not to please the public. The purpose in working it out, is to remove the fault mode from the transmission system. Or, understand the root cause well enough, to plan for future corrections to the scheme.

Since our grid dropped out (a three day outage), there haven't been any since. That might have been fifteen years ago. I like to think, the time spent ensuring the actual root cause was identified, was time well spent. At the time, we did not have a lot of wind and solar, and we still don't have that much solar to speak of.

We have already made corrections to long AC transmission systems, to make them more resistant to solar activity. They aren't bulletproof, as there is always going to be a potential solar event strong enough to do it.

According to my power systems professor, transmission systems are thoroughly simulated, to discover how the system responds to protection events. And the simulations have to be repeated, when changes are made to the setup. This is one way to cover off the embarrassingly simple mistakes. The Russians were the first to use simulation, to improve the dynamic behavior of their grid. Now, the practice is more widespread.

You can't do a good job on the root cause, without instrumentation. And an accurate timebase is essential to good logging.

Paul

Seems quite different.

Paul

None have come forward so far. The hints we have are of a 0.15Hz drop in frequency triggering the shutdown, and two power plants in south-west Spain either going completely offline or with greatly reduced outputs, believed to be solar generators.

I also don't believe that inverters would change frequency under varying loads, so my very wild guess is that the loss of the two plants overloaded the small number of non-renewable generators in the local area, which were the source of the frequency shift. The rest is dominoes...

The problem is that these days, any solution which takes time to reveal is likely to have been fabricated, particularly about a politically sensitive issue, so I'd guess we won't know the real answer until a whistleblower comes forward in a year or two.

I don't think the inverters generate their own 50Hz "clock" and output to that, I think they track what they see from the grid, and then have to output at a fractionally higher voltage than they see as input, in order to "push" their output onto the grid.

All the while looking out for a specified deviation from nominal 50Hz and looking for a specified rate of change of frequency, and disconnecting if either is true.

So all inverters are playing follow-my-leader from each other?

Precisely why we need type approval. If we have that, then then can be rolled off a production line in the same way that cars are.

Any recent design approved for the UK would be a Generic Design Approval (HPR1000 in 2022, AP1000 and ABWR 2017, EPR 2012)

The frequency shift which Andy posted yesterday was between 12:20 and