Most comments at this site tend to have a perspective generally consistent with my own. But sometimes a post will attract comments from people with a very different point of view. That occurred on a post earlier this week titled “Two More Contributions On The impossibility Of Electrifying Everything Using Only Wind, Solar, And Batteries.”
That post and the one immediately preceding it (“Calculating The Full Costs Of Electrifying Everything Using Only Wind, Solar, And Batteries”) had both focused on a particular issue inherent in the project of replacing dispatchable carbon-based sources of energy (coal, oil, natural gas) with intermittent “renewables” (wind, solar). That issue is that, as the intermittent renewables come to provide a greater percentage of electrical generation and as dispatchable fossil fuels get phased out, there is an accelerating need for enormously expensive energy storage to provide the electricity at times when the renewables go quiet. The two posts linked to detailed studies written by four different authors, each of whom had provided a detailed description of their methodology. Two of the four authors even provided spreadsheets, so that a reader who believes the assumptions of the author are wrong can change those assumptions and derive a new cost estimate from the altered assumptions.
The import of all of these studies is that as renewables come to dominate the mix of electricity generation, and particularly as their share of generation goes above 50% and on towards 100%, and fossil fuel backup gets phased out, then the cost of necessary storage becomes far and away the dominant cost of the overall system. Therefore, any meaningful proposal to replace fossil fuel generation with renewables must grapple with this issue.
So what is the solution that the dissenting commenters offer for the problem of increasing need for expensive storage? They don’t offer any at all. Instead, they appear to think that the whole problem can be assumed away or ignored.
The dissenting commenters were three in number, and posted under the pseudonyms “Johnathan Galt,” “GKam,” and “reneawbleguy.” Galt and GKam each posted only one comment, but “reneawbleguy” posted over forty.
The gist of all these comments really comes down to the same thing, namely that the renewables are rapidly becoming cheaper than fossil fuels to generate electricity, if they are not so already, and therefore fossil fuels are a dying industry. Mixed in with this point is a good deal of snide and accusatory language, essentially asserting that anyone who may disagree as to the relative full cost of renewables must necessarily be both ignorant and politically motivated. (e.g., GKam: “More science nonsense from this group of political hacks. . . . Give it up You have already lost.”). Meanwhile, all three fail to deal in any real way with the storage problem inherent in expansion of generation from the renewables.
Here is “reneawbleguy” on the relative cost of fossil fuel electricity generation versus renewables:
Energy costs savings. RE will be cheaper that FF business as usual. 10.43 cents per kw-hr FF 7.81 cents per kw-hr RE. Dollars into our pockets is a clear difference favoring RE. Clear difference.
Money cost savings per person.
No source is cited, but I would agree that approximately these numbers can be found in some studies of relative costs of the renewables versus fossil fuels. But the studies that get these numbers it do so by ignoring the entire storage problem completely.
Similarly, from Galt:
[T]he only consideration to consumers is, was, and always will be “what is the delivered cost to me?” That is neatly quantified in Lazard’s excellent publication providing LCOE.
As I have pointed out on this blog numerous times, the Lazard numbers for “LCOE” (Levelized Cost of Energy) specifically omit any inherent costs of necessary storage. Since the cost of storage is the dominant cost of the all-renewable system, LCOE is the opposite of a “neat quantification” of comparative electricity generation costs, and rapidly becomes completely misleading as the percentage generated from renewables increases beyond 50%.
GKam is even less sophisticated, simply relying on his own personal experience with a home getting its power from rooftop solar panels:
My entire household and both electric cars are powered by the PV system on our roof, as “Galt” can tell you, and it gives us free power having paid back in three years.
GKam does not enlighten us as to how he gets his electricity at night, or overcast days in the winter, or whether he has purchased batteries sufficient to store up power from the summer for use during those long winter nights. If he lives in the United States, it is almost certain that he relies on his local grid — in other words, on fossil fuel backup, with perhaps some nuclear thrown in — for power during those times.
Of the three dissenting commenters, the only one who addresses the storage issue at all is Galt. He asserts, with great confidence, that new battery technologies are coming to make the storage problem go away:
At least two separate technologies, Ambri and Form Energy, will almost certainly have their first large factories up and running within 5 years. Both use common materials (antimony and calcium, iron), both are environmentally safe. Ambri’s battery is 100% recyclable, and in theory may last more than 100 years. Form Energy’s product is likewise 100% recyclable, should cost only 20% that of Lithium Ion, and although the lifespan is not yet advertised it has the potential for similar lifetime of use (simply a “reversible rusting” process).
So the proposal is that a government-mandated total transformation of the entire energy system of our economy should depend on one or another of two not-yet-invented-or demonstrated-at-scale technologies, which may or may not work, and the cost projections of which may be wildly off. Galt does not do any actual numerical calculations. But at a cost of “20% that of Lithium ion” the storage systems he is talking about would still imply a cost of around $100 trillion in Ken Gregory’s spreadsheet, some 5 times current U.S. GDP. Shouldn’t this be acknowledged as a problem? And how can you advocate use of Lazard’s “LCOE” numbers for relative costs of energy sources when those calculations omit a $100 trillion item applicable to wind and solar but not to fossil fuels?
So I say to these three commenters: it’s time to step up your game. Don’t just make unsupported assertions that wind and solar are cheaper. Give us a spreadsheet with a numerical demonstration of how much storage a fully wind/solar/storage electricity system for the U.S. will need, what technology will be used to provide it, and how much that will cost. Without that, you are just dealing in fantasy.
Excellent article. I wish more people understood the issues:
a The invalidity of LCOE when considering intermittent power sources.
b The fact that storage costs dominate once the proportion of RE is significant.
c The fact that batteries are expensive, nasty, and the realities of physics and chemistry will mean technological advancements will be modest.
d The fact that there are very limited geological opportunities for pumped storage.
e The fact that the entire renewable strategy as being executed by the western world sucks.
f The fact that is sucks big time!
Do not forget the laws of economics will not be suspended due to the forced adoption of solar and wind. For example let’s say through whatever method solar and wind comprise 25% of generated electricity thus displacing coal and gas. This would likely lead to a substantial reduction in the price of coal and gas because the supply would be the same, but demand decreased. Will solar and wind be able to compete with natural gas between $1.25 and $1.75 natural gas? Not very well.
Of course, there’s the other options they don’t mention but many of their compatriots believe: Roll back Americans standard of living by a century.
I disagree with the author and most readers of this blog.
No source of energy is 100% dispatchable, it is a matter of degrees. Nuclear is the most available source but it is still well under 100%. Coal is one of the least dispatchable at somewhere in the 60-70% range – far from 100%. Wind varies as in some areas it can be as much or more than coal, while in more marginal locations it is closer to 20-30%. Photovoltaic also varies a lot by location (climate) ranging from 30-60%. Hydropower production also varies a great deal on a seasonal basis and even hourly basis (some hydro plants are used mainly to supply daily peak usage), and typically averages less than 50% annually.
Simplistic analyses such as presented in this author’s posts always ignore demand. Demand is never constant and in fact varies widely by time of day and by time of year, in ways that often actually favor wind and solar. Peak power demand almost always occurs in daytime when most people are awake and working in commercial or industrial activities – when wind and solar produce the most power. Ditto that peak electrical demands occur in summer when solar is at its peak. So logically, use solar and wind for peak generation and thermal plants for base load production.
Indeed most utilities will charge a user much less if they agree to let the utility manage their power supply remotely so as to avoid having to build more peak capacity that is used only a few hours per day.
Power sourcing is therefore NOT a binary, mutually exclusive choice. The issue of matching power supply to demand is never simplistic.
“Everything looks easy when you don’t know what you’re talking about.” Words to live by.
Wind and solar cannot be relied upon solely, but at the same time can still be great economical sources of peaking power, daily and seasonably. The most stable and reliable and economical power generation system is one that is diverse. The optimum mixture of power sources is therefore a combination of coal, gas, nuclear, hydro, wind, and solar, but that mix is also going to be regionalized due to local factors such as climate, available resources, and grid capability.
Another complication due to variability in demand: just because a given power plant is capable of operating at full capacity does not mean that it is actually operated at full capacity. If demand falls below capacity, that plant will only produce what is demanded since there is little to no electrical power storage availability in any existing grid. So when demand falls below capacity the plant produces less power, sometimes by shutting down individual generators or even entire plants.
That is why economical peak generating capacity is so useful to utilities. It costs a huge amount of capital to build thermal power plants relative to today’s cost of wind and solar plants. Building expensive capacity only to see it used only part time is inefficient and wasteful.
Peak demand arises rather late in the day when solar is falling off quite dramatically, to zero in the winter.
For example, last Friday in CA, solar was down 50% by 4:00pm. Demand peaked at 6:00pm and remained high for several more hours.
Wind and solar are useless for peaking power because they can be relied on to produce power when power is needed.
You keep claiming that solar matches well with demand curve, and every time you get shot down by people who actually know what they are talking about.
Solar’s peak output, when it isn’t cloudy is around 1pm to 2pm. In the summer peak demand is around 5pm to 7pm and in the winter peak demand is more like 10pm to midnight.
cannot be relied on
First of all its clear you do not know what dispatchable means. You are confusing it with availability and capacity factor.
Dispatchability is the ability to modulate the output of a power stain to match demand. In the case of wind and solar the only way to do that is by throwing energy away.
Fossil and nuclear can have whatever drives the fire turned down. Coal is of course highly dispatchable, or steam locomotives wouldn’t work.
Taking all those facts into account, what you are really saying is all about availability. which combined with dispatchability gives us the final capacity factor.
So a nuclear power station that is 90% available and is run undisptached will have a capacity factor of 90% also. Like wise wind an solar at around 23% and 15% availability are also run flat out to generate the same capacity factors.
The reason to run the plant this way is to maximize return on capital. The fuel is essentially free.
But when it comes to coal and gas, the fuel is very much not free. If there is so much nuclear or renewable energy that the price of electricity drops below the cost of the fuel, there is no economic justification in running the power station at a loss.
So the low capacity factors on coal and gas are nothing to do with availability, but everything to do with using them as dispatchable generators to match demand and supply.
`
Hydro is a different case, since whilst it is eminently dispatchable, there is generally insufficient rainfall to give year round availability.
People confuse all these issues, technological limitations, and financial ones. Mostly because they don’t really understand either.
LCOE is not “levelized”!
Instead, fossil fuels are encumbered by arbitrary costs or subsidies paid to renewables. e.g., assigning future carbon social costs to fossil fuels.
Alleged renewable costs benefit from those funds taken from fossil fuels plus government mandates for usage or the subsidies given to renewables.
Nor does LCOE contain end of life costs for when solar arrays are dismantled or wind mills fail..
EOL should include bonds paid in advance to remove roads, concrete foundations, every scrap of broken solar cells, etc.
All of those windmill blades that must be taken to the local dump are costs dumped on the local governments.
Then the incredibly rational problem that wind farms get protective contracts requiring local utilities to buy renewable energy over fossil fuel generated energy.
Coupled to this is that renewables do not pay for their interconnects, line stabilizers or the fact that fossil fuels sites must stand idle so they can cover renewable insufficiency or failure immediately.
Renewable installations have a bad habit of failing to live as long as their installers originally claimed.
Another cost ignored by renewables are the costs paid by the destruction of flying wildlife, especially endangered birds or bats.
Stop pretending LCOE levelize costs except as ‘rob fossil fuels or taxpayers to pay renewables’.
Sorry. No space in my garage. Can’t help.
FACEBOOK: “Your message couldn’t be sent because it includes content that other people on Facebook have reported as abusive.”
P’raps if we could manage to build large scale super conductors we could pump renewable electrons into huge super conductor loops and just keep them whizzing around until we need them….just tap into them as required. May I have a cup full of renewable electrons for my Tesla please ?
just dont drop a screwdriver across the terminals..
All I know is that in the UK our energy bills keep going up, with the prospect of eye watering increases in the Spring. This ever escalating cost of energy seems to correlate rather well with the increasd share of unreliables.
Does anyone know who has stolen our wind?
Here in the UK we are having a windless winter. This is very troubling for the nation that Boris has defined as, the Saudi Arabia of wind.
So come on, own up who has taken our wind?
Its lockdown. Normally all the cars rushing around do the trick.
I have been following Wild and Wonderful Off-Grid on YouTube. This lovely couple have built a beautiful home in West Virginia and it is entirely solar-powered and the house and farm runs on 110v. They have 3 children the oldest of whom is just a teenager.
To do this they have 32 panels and 6 (or maybe 8 now) batteries, plus a LPG powered home generator (that they don’t expect to use). This has cost them 10s of 100s of dollars, but since there was no power line part of this is offset by avoding paying for a new connection and they will repay the cost in a few years by avoiding power company charges.
Good for them, it’s a great setup and clearly works, but not many people have the space or spare cash to do that. And it doesn;t include powering an energy-intensive industry.
We installed solar panels when we moved into our new house 3 years ago as we will recoup the investment in about 8 years, but we are a retired couple and we can delay using washing appliances for a day or two until the sun shines and we use excess generated power for an electric heater in winter to reduce our LPG gas usage.
I haven’t been able to justify the purchase of a battery so far, but that may change if it all kicks off in Ukraine…
10s of 1000s, not hundreds…
Storage to cover intermittency is so obviously impossible that it is not a consideration, simple arithmetic will show just how much capacity is required and the extra generation to ensure that the battery is charged. All grid connected batteries are for frequency support, not intermittency.
That aside the renewable enthusiasts seem to ignore that renewables are asynchronous and cannot run a grid. Asynchronysim is like driving a car with no throttle control.
The greater the proportion of renewables feeding a grid the less stable it becomes until eventually if sufficient power is fed in from renewables that it will overwhelm the stabilising conventional generators and power will be lost. And it won’t be easy or quick to restore power when that happens.
Completely true, but how many GreenMinds™ actually understand that, or can even do simple arithmetic?
They deal in touchy-feely qualitative concepts that suit their ’emotional intelligence’ .
We have become a society run by people who have no idea how it actually works.
Renewable energy , which needs massive storage, is more expensive than nuclear power, that does not.
Go figure….
It isn’t. Nuclear is massively more expensive.
The guaranteed payment rate for electricity for UK’s new nuclear reactor at Hinkley is massively above that of any renewable and even current gas price.
The only reason why nuclear is expensive is because opponents like griff do everything in the power to make it so.
In places not ruled by lawfare and where over regulation is kept in check, nuclear is one of the least expensive forms of power.
This has been explained to griff many times, but like his other lies, he doesn’t let reality color any of his opinions.
Yes, intermittency is the problem. When conventional is compared with wind or solar, its apples and oranges. For a fair comparison you have to make sure you are comparing identical products. The key product feature of conventional is that it is consistent, 24 x 7 available, and predictable.
To do a fair comparison to wind and solar, you have to add in the costs of making wind and solar consistent, 24 x 7, and predictable.
Faced with this problem, the renewables lobby has resorted to three tactics.
The first is simple denial. You can read this on Ars Technica, where its comonly asserted by both editorial and commenters (those who are permitted to continue commenting) that in modern economies and grids intermittency simply isn’t an issue. There is never any quantitative justification of this.
The second is to claim that it can be overcome by a big enough grid. You find this in the claims that the wind is always blowing somewhere, so we can just import. Or the sun is always shining somewhere, so we import. This however is totally unsatisfactory as soon as you look at the facts. In Western Europe, for instance, the wind is not always blowing somewhere, especially not in winter. This argument is also never quantified.
The third is more insidious, and is in the very concept of LCOE. The argument is that if you compare LCOE of intermittent and conventional, you find intermittent is cheaper. This has to be combined with some other argument to the effect that intermittency is not an important product feature.
But the concept of LCOE is deficient to start with, despite the fact that its commonly used and widely accepted as a measure of competitiveness.
The reason is, it takes all the power generated over the life of an installation and divides by the total cost. Discount the cash flows to allow for the time value of money, and you get the calculated LCOE per unit of power.
The hidden assumption here is that it doesn’t matter when the power is generated. But of course, that is exactly the problem with wind and solar. Peak demand in the UK, as a for instance, occurs between 5pm and 10pm on dark cold winter evenings. And this is the period when, for a week or so every winter, there is a blocking high and so minimal generation.
But the assumption in the LCOE method is that this does not matter to value. It doesn’t matter when you generate it, its value is assumed to be the same. Whether its in hot breezy and sunny July, generate all you want, whether its needed or not, and it will count as raw megawatts generated. Fail to generate in cold dark calm Jan or Feb, and this will not enter into the calculation either.
I have sometimes trying to explain this used a lettuce example. Its like we are s supermarket trying to meet lettuce demand. One supplier ships whenever they feel like it, sometimes a truckload on Monday evening, sometimes none for a week or two. Another delivers a constant amount twice a week.
We have to throw out half what the first supplier delivers because it goes off before sale, owing to mismatch with demand. We sell just about all of the second supplier’s deliveries.
The first supplier now says they want to be paid the same as the second for every lettuce they deliver, delivered whether we want it and can sell it or not. Deliver a month’s supply on Monday morning, get paid, then deliver some more when they can,maybe in a few weeks.
This is the fundamental intellectual dishonesty at the heart of LCOE. Make the products comparable in features, and intermittent makes no sense except in very special circumstances. I can imagine for instance that in super hot climates with cool nights, solar to power AC might fly financially. As soon as you get hot nights however, no way.
The LCOE calculations quoted, such as Lazard, tend to only use 30 years as the measurement period, and they then claim that both solar and wind systems will survive for these 30 years. But if you actually compare energy sources of their full lifetimes, such as nuclear with over 60 years, then you have to factor in the renewable replacement costs. This would greatly alter the cost comparisons.
It’s quite dishonest to claim that wind or solar will last for 30 years. Real world data shows actual lifespan for both is much shorter. Especially for offshore wind.
In addition, they are assuming that solar will continue to produce first day power, for it’s entire lifespan. That is also demonstrably wrong.
I remember reading somewhere that the US Energy Information Administration also use the 30 years measurement period for wind and in their comparisons with other forms of power generation use the same 30 year life even though, as you say, the actual operating life of coal, gas and nuclear is far greater than 30 years.
“Hydrogen generation from low-cost renewables at $25/MWh with a capacity factor of 50% yields a cost of $1.70/kg of hydrogen produced. Storing this hydrogen underground will add about another $0.30/kg, thus the hydrogen costs $2/kg. If this hydrogen is used to generate power, the resulting cost is $100 to $200/MWh. In ideal conditions (e.g. a CCGT turbine at 60% utilisation), the cost is $100/MWh, while simple-cycle turbines at 25% utilisation would deliver power at $200/MWh.”
https://www.powermag.com/how-much-will-hydrogen-based-power-cost/
Fantasy. You’re not going to get $25/MWh renewables anyway. Use to make hydrogen either adds to demand instead of supplying it when there is no surplus production, so the cost becomes what other generation you fire up to meet demand, or it relies on highly intermittent and variable amounts of surplus, which are never going to reach 50% capacity factor for any appreciable volume. See this chart
https://datawrapper.dwcdn.net/nZM72/1/
The exact numbers is not the point. 25 or 30 $ doesn’t matter much. We don’t need to store all of the RE produced. Mentons posts are just red herrings in the storage debate. I doubt anybody ever suggested using Li batteries for grid scale seasonal storage.
I think you’re saying that storing excess electricity production as hydrogen and then using the hydrogen to run OCGT to fill in gaps between demand and supply will cost, for the OCGT, 8x as much as the initial excess electricity.
Hydrogen CCGT at 4x the cost of the electricity used to produce the hydrogen doesn’t seem economic, due to the time it takes to ramp up. Perhaps it could be scheduled in conjunction with solar PV, but wind doesn’t have the cyclic nature of day and night.
A solar PV/OCGT/CCGT hybrid seems an interesting alternative to molten salt solar thermal for shorter terms where seasonal factors don’t come into play.
The order of magnitude numbers are the point. Firstly, the scale of the storage requirement in terms of redeliverable energy, which is substantial and far beyond most estimates of covering a handful of days of low renewables production that are typically used in these evaluations. And secondly in the real costs and round trip efficiencies that drive the rest of the equation. I calculated that to have supplied the UK during 2021 from wind generation there would have needed to be over 50TWh of hydrogen storage that would have had to be already over half full at 1 January. At 60% efficiency for electrolysis and 60% efficiency for a CCGT to burn the hydrogen to produce power again you are down to 36% round trip efficiency, which already severely taxes the economics.
It is always cheaper to look at solutions with proper dispatchable backup rather than storage, and the trade off between storage and simply spilling excess power greatly favours the latter, but you eventually reach unavoidable levels of storage or proper backup.
I’ve done the numbers. Please, no battery red herrings designed to fool green ingenues.
The big problem is that there is no such thing as low-cost renewables, and never will be.
Synfuel. 15 eurocent/kwh by 2030
https://www.agora-energiewende.de/en/publications/the-future-cost-of-electricity-based-synthetic-fuels-1/
Superb reply to reneawbleguy et. al, Mr. Menton
May I suggest 2 enhancements to future versions?
1) Make the following point. INTERMITTENT renewable electricity differs fundamentally from the RELIABLE electricity to which we are accustomed. Intermittency REDUCES the value of renewable electricity to electricity users. Furthermore, intermittent generators need 3-4x more transmission line capacity than thermal & hydro generators. Thus, straight comparison of LCOEs is wrong.
2) Suggest creating a new electricity service category–SOLELY solar & wind 365. Lack of electricity at night, insufficient electricity on calm days, and huge, frequent fluctuations most days & nights may smarten up renewable enthusiasts.
As for renewables being cheaper than fossil fuels, the proof is in the deployment. Germany is highly reliant on renewables but has electric rates 2 or 3 times the US, which uses much less renewables. California is highly reliant on renewables and is in the top 5 states for consumer electric rates, as well as suffering brownouts or rolling blackouts during peak usage times. This type of data belies the “renewables are cheaper mantra”.
Much of the German electricity cost is tax. And the green levy component on it will fall this year. Plus very many German households have solar and excellent insulation. German households use less electricity than US ones.
German households use less electricity than US ones precisely because electricity is so much more expensive.
Can you demonstrate that German taxes have increased this dramatically in recent years? Or is this just another lie that you have been told to push?
Gkam and Galt post regularly on another site I read. Gkam has repeated his personal story a thousand times. Galt is reasonable until it gets to storage. He than repeats the Ambri and Form story. Don’t waste your time responding to them.
And if you wish hard enough, Tinkerbell will come back.
Baby Boomers were taught this as children. It’s no wonder they still believe in fairy dust and unicorn rainbows.
It is more than storage being ignored. FF incurs more cost as a large portion of generating assets need to be held on standby in case the wind stops blowing. So the very existence of the unreliable renewable generating sources on the grid drives up the costs for the reliable fossil fuel sources because they are only partially utilized and so their rates have to incur more capital and overhead costs. Without the RE on the grid, under a more full utilization of assets (think revenue per dollar asset) their costs very likely could be just 6-7 cents.
Over the years, I have noted that many environmentalists hide their true goals. We are told one (usually reasonable) thing when the hidden goal is something else entirely.
I believe that power storage is one of those things. Currently power supply is varied (dispatched) to follow demand. Wind and solar are capable of doing this, requiring storage to provide/absorb the difference.
If, however, demand is forced to follow supply, the whole problem of storage goes away. I believe that this is the true (unstated) goal of renewable energy.
Mandating dependence on nonexistent technology?
Echoes of Nonqawuse
https://en.m.wikipedia.org/wiki/Nongqawuse
I was wondering this holds true for all wind turbines. When temps drop below -22°F(-30°C) turbines automatically shut down (for safety) and become a net consumer of electricity to maintain a minimum operating temp.
https://www.americanexperiment.org/it-wasnt-very-windy-this-morning-and-thats-a-problem/