This group brings together the best thinkers on energy and climate. Join us for smart, insightful posts and conversations about where the energy industry is and where it is going.

Post

California secretly struggles with renewables

image credit: © Adam Webb | Dreamstime.com

California has hooked up a grid battery system that is almost ten times bigger than the previous world record holder, but when it comes to making renewables reliable it is so small it might as well not exist.

The new battery array is rated at a storage capacity of 1,200 megawatt hours (MWh); easily eclipsing the record holding 129 MWh Australian system built by Tesla a few years ago. However, California peaks at a whopping 42,000 MW. If that happened on a hot, low wind night this supposedly big battery would keep the lights on for just 1.7 minutes (that’s 103 seconds). This is truly a trivial amount of storage.

Mind you this system is being built to serve just Pacific Gas & Electric. But they by coincidence peak at about half of California, or 21,000 MWh, so they get a magnificent 206 seconds of peak juice. Barely time to find the flashlight, right?

There is no word on what this trivial giant cost, since PG&E does not own it. That honor goes to an outfit called Vistra that does a lot of different things with electricity and gas. But these complex battery systems are not cheap.

This one reportedly utilizes more than 4,500 stacked battery racks, each of which contains 22 individual battery modules. That is 99,000 separate modules that have to be made to work well together. Imagine hooking up 99,000 electric cars and you begin to get the picture.

The US Energy Information Administration reports that grid scale battery systems have averaged around $1.5 million a MWh over the last few years. At that price this trivial piece of storage cost just under TWO BILLION DOLLARS. At 103 seconds of peak storage that is about $18,000,000 a second. Money for nothing.

Mind you the PG&E engineers are not that stupid. They know perfectly well that this billion dollar battery is not there to provide backup power when wind and solar do not produce. In fact the truth is just the opposite. The battery’s job is to prevent wind and solar power from crashing the grid when they do produce.

It is called grid stabilization. Wind and solar are so erratic that it is very hard to maintain the constant 60 cycle AC frequency that all our wonderful electronic devices require. If the frequency gets more than just a tiny bit off the grid blacks out. Preventing these crashes requires active stabilization.

Grid instability due to erratic wind and solar used to not be a problem, because the huge spinning metal rotors in the coal, gas and nuclear power plant generators simply absorbed the fluctuations. But most of those plants have been shut down, so we need billion dollar batteries to do what those plants did for free. Nor is this monster battery the only one being built in California to try to make wind and solar power work. Many more are in the pipeline and not just in California. Many states are struggling with instability as baseline generators are switched off.

There is even an insane irony here, one that is perfect for Crazy California. This billion dollar battery occupies the old generator room of a shut down gas fired power plant. Those generators used to make the grid stable. Now we are struggling to do it.

Of course no one at PG&E or Vistra says publicly that this monster battery is there to keep renewables from trashing the grid, not to back them up. One wonders if the California Public Utilities Commission knows this? The big question is why is the rate paying public not told? Or the press? There is really a very expensive hoax here.

While on this topic, let’s ask what it would actually cost to back up wind and solar with batteries. This depends a lot on local climate. How often the wind does not blow hard for example. Wind generators need about 10 mph just to start and more like a sustained 30 mph for full power.

Multi-day heat waves are often periods of very low wind, combined with a maximum need for power. A nasty combination. So my rough rule of thumb is that you need storage of 7 days times peak need.

California peaks at 42,000 MWh and 7 days is 168 hours so using this rough rule we would need about 7 million MWh of batteries. This makes 1200 MWh truly trivial. Then at $1.5 million a MWh we get an astounding 10.5 TRILLION DOLLARS, just for the batteries to make renewables reliable.

The scam is breathtaking, and not just in California. Nationwide we are spending untold billions of dollars trying to keep the erratic nature of renewables from crashing the electric power system. But these efforts are routinely portrayed as storage for when renewables do not run. Stabilization is the opposite of storage. We are being lied to about renewables.

David Wojick, Ph.D.'s picture

Thank David for the Post!

Energy Central contributors share their experience and insights for the benefit of other Members (like you). Please show them your appreciation by leaving a comment, 'liking' this post, or following this Member.

Discussions

Spell checking: Press the CTRL or COMMAND key then click on the underlined misspelled word.
Bob Meinetz's picture
Bob Meinetz on Jan 24, 2021

David, I couldn't agree more: we are, in fact, being lied to about renewables.

Two days ago the California Public Utilities Commission (CPUC) scuttled an evidentiary hearing in their attempt to prevent California's blackouts of last summer from occurring again in 2021. Apparently, CPUC doesn't feel the "breathtaking scam" being perpetrated on consumers should be public knowledge. The problem, a recent report claims, is not that we're wasting money on batteries, but that we're not wasting enough. Spending $billions more on batteries, while building more "resource adequacy" (gas plants), has been deemed the quickest and most economical  route to zero carbon emissions.

Go figure.

Hard to blame a commission beset by scandal for wanting to keep a low profile. Only three years ago its president was being investigated by none other than Kamala Harris, in her role as CA Attorney General. With the possibility it might thwart her political aspirations, however, her investigation was dropped.

As you correctly note, California's grid batteries serve to fix the problems caused by erratic solar and wind generation. But to independent power producers (IPPs), they're also useful for saving cheap electricity generated overnight at gas plants, for resale the next day during peak consumption - a travesty that offers zero benefit to consumers. Not only are the sky-high prices reflected in electricity rates, but batteries waste anywhere from 10-25% of energy stored in system losses - as heat - and every cubic foot of gas used to generate it is billed to ratepayers, too, at a markup.

Peter Farley's picture
Peter Farley on Jan 26, 2021

A combined cycle gas plant operating in the cool of the night and into the early morning will have an efficiency of around 55-60%. An open cycle gas turbine running at part load just in case on a 100F day can have efficiency as low as 30% or worse. Using the CC plant to charge the batteries is cleaner and cheaper than running gas peakers, so again you are exactly wrong using batteries and hydro to flex so that CC gas plants can run at optimum efficiency saves customers money and reduces total system investment 

Bob Meinetz's picture
Bob Meinetz on Jan 26, 2021

You are using batteries that are 100% efficient, are you? What technology would that be?

Peter Farley's picture
Peter Farley on Jan 28, 2021

Standard battery technology. A 55% efficient CC plant charging an 85% efficient battery on the site of the CC plant has a system efficiency of 45%, far better than an OC plant. In addition reduced ramp rates on the CC plant reduces maintenance costs.

Further nearing peak times many OC plants are started "just in case" of a rapid load rise or trip on another generator, because gas turbines can't start under load and they aren't stable at less than around 30% load you can have three plants running inefficiently at 30% rather one at 90% to supply the near peak load. If batteries are used to supply short term needs, then the OC plants can be turned off until there is a genuine hour+ need, again saving fuel and maintenance cost at the OC plant. Once the gas turbines do start, they can run in combination with the batteries to meet peak load and then as load moderates the turbines run can continue at near full load for an extended time to recharge the batteries, thus improving the average run time and therefore maintenance costs per MW and reducing fuel use per MWh by running at 60-90% of rated power for some hours rather than low power for 1-3 hours high power for an hour or two and then ramping slowly down to off.

 I hope that explanation is helpful 

john Liebendorfer's picture
john Liebendorfer on Jan 24, 2021

This article is so full of half truths and outright lies, I am disappointed to see it published on such a reputable site.  There is an “alternative fact” in every paragraph.

First a couple of general comments.

-  Climate change is real and we must get stop using fossils fuels or California will LITERALLY burn up as it is already doing.

-  California is the 6th largest econonmy in the world.  Their efforts, regardless of how incidental,  will encourage others to follow suit.

The Specifics. 

1- The Battery system described is not intended to back up the grid if it crashes.  Rather it is intended to even out the supply and demand on a given day.2- This is a small system, but it is the first step, not the last step.  In the coming years these sorts of projects will contribute to stabilizing the grid under more serious conditions.3- This will not impact cost to electric consumers.  Rate payers are not paying for this systems, private investors are.  4- States batteries average cost over the last few years are $1.5 million per Mwh.  But battery costs are rapidly declining - 13% in the last year.  So an historic average cost is much higher than the marginal cost of the next system.  In 2020 the average cost was $1.37 Mwh.  Next year it will be less.5- States the $1.5 million per Mwh cost is not cheap, Compare that to the Vogle Nuclear plant in George costs of about $10 million per Mwh.  With utility scale solar installing at under $1 million per Mwh, solar with battery backup at $2 to $2.5 million Mwh is 1/4 of the price of new nuclear.6- He indicates that building these battery “farms” with 99,000 modules is a challenge beyond our capability.  Yet 2017 estimates say Google has 2.5 million servers spread across 30 or more server farms around the world. (I found the number instantly by a google search)7- Claims that solar and wind mess with the grid stabilization such as maintaining 60 cycle frequency.  That is patently false. In fact the power coming out of solar invertors is far cleaner power than the existing grid power and actually helps clean up the grid.8- Claims the the spinning rotors of fossil  and nuclear plants absorb the fluctuations of renewables.  AND most of those plants have been shut down.  Again patently false.  85-90% of electricity still comes from coal, gas, nuclear, and nuclear.        The entire article is just fear mongering   


 

 

Roger Arnold's picture
Roger Arnold on Jan 25, 2021

This article is so full of half truths and outright lies, I am disappointed to see it published on such a reputable site.  There is an “alternative fact” in every paragraph.

I respectfully disagree. The article is actually pretty accurate in terms of facts. One may not like the way the author spins the facts, but you're mostly taking issue with things he doesn't actually say.

My impulse to to defend the article may owe something to the fact that the author at least understands the difference between MW and MWh. He uses the terms correctly according to context. I. know that's a very low bar, but it's astonishing how many people flunk it.  A friend of mine had a good simile. Confusing MW and MWh is like confusing meters with meters per second. "How tall are you, in meters per second?"

The Specifics. 

1- The Battery system described is not intended to back up the grid if it crashes.  Rather it is intended to even out the supply and demand on a given day.

The author didn't say it's intended to back up the grid if it crashes. He said it's intended to prevent the voltage and frequency variations that variable renewables would otherwise cause from crashing the grid. He's right, and he's actually saying pretty much the same thing you are: responsive battery banks smooth out short term changes in supply and demand and buy time for slower responding dispatchable power generators to ramp. The battery banks don't have nearly the capacity that would be needed to fully even out supply and demand over the course of a full day. The system is still very much dependent on the availability and operation of dispatchable generation -- which usually means fossil fueled. That's the point.

2- This is a small system, but it is the first step, not the last step.  In the coming years these sorts of projects will contribute to stabilizing the grid under more serious conditions.

One would hope so. But it's important to recognize the magnitude of the gulf between where we are now and where we need to be. We won't get there through modest annual additions to battery storage and renewable energy capacity. If we hope to get there quickly at a reasonable cost we'll have to accelerate deployment of other technologies. We'll need either vastly cheaper large scale / long term energy storage, or vastly larger amounts of zero carbon emission dispatchable generation.

I don't think I'll try to address the other points of your specifics individually. I've already bashed you about the head and shoulders enough for one reply, and I do in fact have some sympathy for your comments. But in the future, please do try to avoid anything as inane as your point 4 -- that rate payers are not paying for the systems (variable renewables and battery storage) because private investors are paying for them. It's hardly a coincidence that in Germany and California, the country and the "almost country" with the most aggressive programs for deployment of variable renewables in the western world, ratepayers "enjoy" the highest priced electricity of any non-island nation anywhere in the world.

Joe Deely's picture
Joe Deely on Jan 25, 2021

Is this really Roger?? or his evil twin?

Not gonna waste my time debating this garbage article.... but will point out one small thing.

You said:

My impulse to to defend the article may owe something to the fact that the author at least understands the difference between MW and MWh... I. know that's a very low bar, but it's astonishing how many people flunk it.

Author wrote:

Mind you this system is being built to serve just Pacific Gas & Electric. But they by coincidence peak at about half of California, or 21,000 MWh,

Just to prove it wasn't a typo later the author said:

California peaks at 42,000 MWh 

Looks like somebody is coming in under that "very low bar". You might even ask - 

"What's your instantaneous peak in MWh?

 

Bob Meinetz's picture
Bob Meinetz on Jan 25, 2021

At the beginning of the article David writes

"However, California peaks at a whopping 42,000 MW."

showing that he does indeed understand the subject - and showing your basis for labeling it a "garbage article" amounts to two typos.

Overall the author is spot-on: battery storage for electrical power, not ancillary services, could power the CAISO grid for a period measured in seconds. As far as "backing up" renewables, it's useless.
 

Peter Farley's picture
Peter Farley on Jan 26, 2021

The article is garbage, apart from confusing MW with MWh, he assumes there is no hydro, no solar, no geothermal and that the wind is actually zero for two weeks. Wind in some areas might be very low for days at a time but it is never zero across the state for two weeks. 

Then he claims that wind turbines need 10 mph winds to start when modern wind turbines are operating at 7mph. At ground wind speed of 10 mph a modern turbine like a GE 2-127 on a 114 m tower would be seeing almost 6m/s wind speed and running at about 160 kW and would reach maximum power at a ground wind speed of 18-19 mph. The importance of this is that wind speeds at or above 18-19mph are more than twice as common as wind speeds above 36mph.

Then he has it completely backwards about stability of voltage and frequency. 

In summary almost every point he makes is either flat out wrong or a distortion   

Roger Arnold's picture
Roger Arnold on Jan 27, 2021

Is this really Roger?? or his evil twin?

It was my evil twin (ET). I try to keep him leashed, but he sometimes gets loose. Especially if I'm in a particularly grouchy mood, or angry about something unrelated that I feel helpless to do anything about.

A couple of things you should know about ET, the next time you encounter him. He's reactive, hair triggered, and sometimes sloppy (e.g., failing to notice Dr. Wojick's bloopers with "peak MWh" in his 3rd paragraph, while being annoyed by John L's manglings in his points 4 and 5.) ET doesn't like to do homework, if he thinks he already knows the answer. OTOH, he never lies, or intentionally tries to mislead. He and I share the same heart, the same values, and the same technical knowledge. Also, FWIW, the same stiff, aching joints.

ET and I are both balance-seekers. Natural contrarians. If somebody takes a strong position on something, we automatically start looking for holes. There's plenty to criticize in Dr. Wojick's article. If John L.'s reply hadn't been so over the top, I'd have been in there with my own critiquing. But for the most part, Wojick is spinning the facts, not misstating them. There's a difference, and it needs to be recognized.

 

Matt Chester's picture
Matt Chester on Jan 27, 2021

But for the most part, Wojick is spinning the facts, not misstating them. There's a difference, and it needs to be recognized.

Something to always keep in mind, wise words! Thanks Roger/ET!

Joe Deely's picture
Joe Deely on Jan 27, 2021

Pretty close to my evil twin... 

 But for the most part, Wojick is spinning the facts, not misstating them.

As far as facts... let's take a quick look at one "fact".

 The US Energy Information Administration reports that grid scale battery systems have averaged around $1.5 million a MWh over the last few years. At that price this trivial piece of storage cost just under TWO BILLION DOLLARS

Here are Vistra's storage projects.

  • Upton 2 (10 MW/42 MWh) – online December 2018
  • Moss Landing – Phase I (300 MW/1,200 MWh) – online December 2020
  • Moss Landing – Phase II (100 MW/400 MWh) – expected online by August 2021
  • Oakland (36.25 MW/145 MWh) – expected online 2022 
  • DeCordova (260 MW/260 MWh) – expected online 2022

Let's say about 2,000 MWh of storage. Based on the "fact" above this would be a $3B investment. Phase I and II of Moss Landing would be $2.4B alone. Vistra is also developing 1GW of solar. Let's say another $1B. So total $4B investment in storage/solar.

This is from Vistra 2019 Annual report.

Moss Landing — In June 2018, we announced that, subject to approval by the CPUC, we would enter into a 20-year resource adequacy contract with Pacific Gas and Electric Company (PG&E) to develop a 300 MW battery ESS at our Moss Landing Power Plant site in California. PG&E filed its application with the CPUC in June 2018 and the CPUC approved the resource adequacy contract in November 2018. At December 31, 2019, we had accumulated approximately $64 million in construction work-in-process for this ESS. Under the contract, PG&E will pay us a fixed monthly resource adequacy payment, while we will receive the energy revenues and incur the costs from dispatching and charging the ESS.

Two items there -  

1) spend in 2019 on Moss Landing project was $64M. Let's see what spend turns out to be in 2020. Guidance for spend on storage AND solar projects in 2020 was $377M. Long ways to go before we get to the $4B as needed to support the "fact" above.

2) Vistra will get some capacity payments from PG&E but is also paying off this storage investment by charging and dispatching this battery on CAISO. Would love to see somebody "pencil out" Vistra getting their supposed $2.4B investment back using this business model.

Facts? Give me a break.

Bob Meinetz's picture
Bob Meinetz on Jan 28, 2021

"Let's see what spend turns out to be in 2020."

Or,  "Let's budget new projects ahead of time, instead of handing over customers' money to storage developers then waiting to see what they come up with."

"Would love to see somebody 'pencil out' Vistra getting their supposed $2.4B investment back using this business model."

No doubt Vistra has already pencilled in exactly how they'll extort ratepayers for the full price of their investment.

Roger Arnold's picture
Roger Arnold on Jan 29, 2021

True, $1.5 million / MWh is a gross overestimate for the cost of large grid-scale battery storage projects going forward. OTOH, it illustrates fairly well what I was trying to get at with the distinction between "spinning the facts" vs. misstating them.

Dr. Wojick's statement, "The US Energy Information Administration reports that grid scale battery systems have averaged around $1.5 million a MWh over the last few years." is essentially correct! A quick google search on "EIA estimate average cost of grid-scale battery storage" turned up this report as the first hit. It doesn't include the exact wording that Wojick used, but it includes a nice graph of installed battery storage costs for the years 2015 - 2018. The graph shows both the 25th to 75th percentile cost range for those years and the capacity-weighted average. The capacity weighted averages for 2015 - 2017, taken together, do come out close to $1500 per kWh ($1.5 million / mWh).

I don't know if that particular report was the source for Dr. Wojick's statement about average costs, but the whole point of it was that costs had fallen nearly 70% from 2015 to 2018. The capacity-weighted average for 2018 was only $625 / kWh. Some of the drop would have been from a decrease in the specific cost of the battery cells used, but I suspect that an equal factor was that projects have been getting larger. "Balance of system" costs would be proportionately less important. 

None of that context about falling costs is evident in Wojick's flat statement. Its absence renders a statement that may be factually correct -- as far as it goes -- nonetheless highly misleading. Cherry picking data and omission of context are two ways to "spin the facts" without actually misstating them (aka outright lying).

My admonition about the need to distinguish the two wasn't intended to suggest that one was OK while the other isn't. I just think that people should be aware of how differently the same facts can be seen, when viewed through different bias filters.

Bob Meinetz's picture
Bob Meinetz on Jan 30, 2021

"True, $1.5 million / MWh is a gross overestimate for the cost of large grid-scale battery storage projects going forward."

EIA notes the 2018 installed cost for large, grid-scale, Li-Ion battery storage projects has dropped by 70% since 2015. Historically, it's approaching an asymptotic cost of ~$500M / MWh.

What's the basis for your belief if will drop another 70% in any appreciable timeframe?

Bob Meinetz's picture
Bob Meinetz on Jan 25, 2021

Thank you, Roger. You've earned the title of "Evil Twin" from a renewables supporter -  you must be doing something right.

Joe Butzer's picture
Joe Butzer on Jan 24, 2021

I am also disappointed in this article. This article is not intended to educate people but only to intimidate and scare them. There is no way I can argue any of the points made and I guess that is the point in todays society. Anyone can say anything, quote a few numbers and look like an authority only delivering the truth. I work everyday with the electrical grid economically dispatching generation and making daily decisions regarding grid reliability. 

The simple truth is that our grid is run and driven by economics and the market. Everyone offers a product into the market and everyone purchases what they need from the market. The market dynamics today make this battery economically viable. The battery pack will pay for it's self in months due to how it is offered into the market and the revenue it can generate. Not even considering the social benefits of renewable energy, but based on todays market and economics batteries are a great investment. Our system is dynamic and continually changes but we need encouragement for batteries and other renewables not crazy analysis that do not reflect todays economic realities of the electrical market and how products are offered into the market.

Bob Meinetz's picture
Bob Meinetz on Jan 25, 2021

Joe, you're missing the point. Grid scale batteries are being sold to the public as if they can fill in for a solar farm overnight when the sun isn't shining, or cover for a wind farm for an extended period of calm. As though they're capable of providing power electricity, not simple frequency, voltage, or phase adjustments. But they aren't - not even close.

Imagining a grid completely powered by renewables and batteries requires ignoring some inconvenient truths:

  1. To power the CAISO grid with batteries, for one day of cloudy, calm weather, at average consumption, would require battery storage capacity costing $428 billion - twice California's entire state budget.
  2. What's powering California's grid while the batteries are being charged? To have 100% clean electricity, we'll need to overbuild wind and solar farms by 30% (cha-ching).
  3. At the beginning of the second day of cloudy, calm weather, the entire state would go dark.
  4. How much storage would we need? Enough for three days, or five days? Because allowing the grid to go down is not an option, there can never be enough battery storage to power an electrical grid.
  5. Do we need to power the grid at higher levels of consumption, even up to double that of an average day? Of course - now we're up to $856 billion/day, or four times California's budget.
  6. Lithium-ion batteries have a limited lifetime - they would need to be replaced every 7-10 years.
  7. Over time, if its grid batteries last eight years, California would need to spend half of its budget every year on grid batteries - as much as all public education combined.

Ain't ever gonna happen - and that's the point.

"This article is not intended to educate people but only to intimidate and scare them. There is no way I can argue any of the points made and I guess that is the point in todays society."

Don't feel bad, no one can argue any of the points made. Scary or not, David is correct - renewable electricity is a scam of breathtaking proportions.

Peter Farley's picture
Peter Farley on Jan 26, 2021

Firstly he claims with no basis in fact that you need two weeks at peak demand when peak demand rarely last more than two hours, so even if two weeks storage was needed it would at worst be two weeks demand at about 2/3rds of peak hour. 

Then he implies that during that two week period there is zero wind, zero solar and zero hydro and zero geo-thermal or waste to energy or imports from other states. Again a completely untenable position.

Finally he implies that there will be no cost reduction in storage or alternative long duration storage such as flow batteries, pumped hydro, CAES, or hydrogen.

 It is just ignorant scare mongering  

Bob Meinetz's picture
Bob Meinetz on Jan 26, 2021

"...so even if two weeks storage was needed it would at worst be two weeks demand at about 2/3rds of peak hour..."

OK Peter, let's calculate how much battery storage capacity we would need using your assumptions for a worst-case scenario - after all, the grid can't go down, can it?

22,000 MW x .66 = 14,520 MW
14,520 x 336 (hours in two weeks) = 4,878,720 MWh

We'd need 4,878,720 megawatthours of battery storage capacity to power California for two weeks.

Cost of battery storage (EIA - 2018): $625,000/ MWh
4,878,720 x 625,000 = $3.1 trillion

Again:  powering a grid with renewables + batteries makes no sense at all - in cost, physics, emissions, or anything else. A pipe dream.

Peter Farley's picture
Peter Farley on Jan 26, 2021

You don't need two weeks of battery storage or even a whole day. Why do you continue to ignore solar, hydro, geothermal, waste to energy, thermal storage and residual residual wind. Net result is battery storage worst case is about 24 hours. A large fraction of that will be behind the meter and/or in EVs. the average EV has 2-7 days storage on board. Then there is grid controlled hot water and cooling. A typical domestic hot water heater properly configured can store three days of hot water. Similarly commercial and industrial cooling systems using glycol or ice can economically store days worth of energy.    

Therefore grid scale battery storage will be limited to six to twelve hours. To be really conservative use 20 hours at 30,000 MW about 1/10th of your figure

And then you use completely out of date battery storage costs. Latest deals are coming in at $200-250,000 per MWh and falling at 16% per year so by 2025 a battery installation at an old power plant or existing hydro installation or substation where existing transformers and switchgear can be used, will be coming in at $100-120,000 per MWh.  Using the higher figure the cost of utility battery storage will be in the order of $72 billion, or three plant Vogtle's.

The difference is that peak output of the nuclear plants would be 6.6 GW. If the battery power systems are configured as 12 hours at full output or 20 hours at 60% output the peak summer output will be 50 GW eight times that of the nuclear plant. 

Bob Meinetz's picture
Bob Meinetz on Jan 27, 2021

"Why do you continue to ignore solar, hydro, geothermal, waste to energy, thermal storage and residual residual wind..."

Peter, I ignore waste to energy, thermal storage, "residual residual wind" (whatever that is), tidal energy, flywheel storage, hamster wheels, grid-controlled hot water, mind-controlled zero-point energy, etc., because they're pixie dust.

I ignore geothermal because it's severely limited, and hydropower because it isn't dependable either. Both are limited by location and non-scalable.

"A large fraction of that will be behind the meter and/or in EVs,

Sounds like your system will only work for rich people. I was kind of thinking clean, cheap energy should be available for everyone.

"...the average EV has 2-7 days storage on board."

And your EV is powered by...what?

"Net result is battery storage worst case is about 24 hours..."

(one paragraph later...)

"Therefore grid scale battery storage will be limited to six to twelve hours."

Dropping like a rock! By now we shouldn't need any storage, right?

"And then you use completely out of date battery storage costs. Latest deals are coming in at $200-250,000 per MWh..."

Nonsense. The latest numbers for installed storage (from the impartial EIA, not the renewables echo chamber) is $650,000/MWh.

Peter Farley's picture
Peter Farley on Jan 28, 2021

So now you agree that

1. The estimate of storage costs and duration in the article is wildly exaggerated. i.e $650,000/MWh last year not $1,500,000.

According to the statutory accounts of Neoen the 129 MWh battery in South Australia in 2017 cost A$90m (<US$60m at the time) i.e US$465/MWh. According to Elon Musk the current price for battery storage due to the use of ever cheaper batteries and 3 MWh Megapacks instead of the 250 kWh Powerpacks used at Hornsdale has reduced prices to US$200-300/MWh. Who knows the current cost better, the manufacturer quoting the next jobs to be installed or the EIA surveying 12-18 month old data

2. That the quantity of storage is also vastly exaggerated. You ignore hydro why because its output varies why?, so does nuclear, gas and every other power source. Even in France nuclear power was down 40,000 GWh last year compared to the year before.  I agree waste to energy, landfill gas, geothermal will be small contributors but they are dispatchable and in the hypothetical wind drought they can supply about 5% of demand while hydro can supply almost 30% for a few days. There is no reason why overbuilt solar cannot supply 50% of demand because like South Australia, it can supply all the demand for 4-5 hours per summer day and then additional solar can be used to heat water, power municipal water transfers, make ice etc.  In case you suggest that it can't supply 50% of power in winter, that may be true but it is not relevant to the article which is all about summer peaks.

Residual wind is the wind energy generated in so called wind droughts. There are three regions where hourly wind energy data is easily accessible to the public. The UK, Germany and Australia and more specifically South Australia. All of those regions have wind droughts but none have zero wind ever. It may be down to 2% of capacity for three or four hours half a dozen times per year and it may be down to 20% of annual average output for three days or 40% for two weeks but if California increases its wind generation capacity per square mile to that of Germany today, average output from wind will be about 500,000 GWh per day. In the worst two weeks that will be 200,000 GWh per day, reducing the required storage by  67 million GWh 

3. In South Australia night-time hot water storage increased demand by about 200 MW out of average night-time demand around 1.200 MW. That can all be shifted to the middle of the day 

4. Rich people charging their home batteries from rooftop solar or off peak electricity reduce peak demand and therefore the cost of infrastructure for everyone else.

By buying expensive EVs and solar panels they drive volumes up and prices down so that now the five year cost of ownership of a new VW ID3 is lower than the cost of a new gas powered VW Golf. 

Once you get rid of your silly net metering (I think we agree on that) and have time of day tariffs that reflect wholesale prices, rooftop solar acts to reduce wholesale prices as it has in Australia and Germany, so whether or not you have your own battery or solar panels power prices can be and are being reduced.

 

Peter Farley's picture
Peter Farley on Jan 25, 2021

I am not sure whether this is supposed to be a joke or is just unbelievably misinformed. its random swapping of MW for MWh seems to indicate the latter. For example the peak power demand is 42,000 MW not 42,000 MWh. Peak day energy use is more like 30,000 MWh

1. This plant represents just under 1/4 of the utility batteries being installed in California by the end of this year and California is not phasing out its conventional power plants instantly it has many years to provide the resources it needs to replace FF plants

2. California has 14 GW of its own hydro capacity and can import about 3 GW continuously from the Pacific Northwest so why would anyone expect a single battery to provide all its power or energy needs. It is like assuming you have to build enough gas stations so that everyone can fill up at the same time or even more directly that a single gas turbine plant would solve all its problems.

3. In a hypothetical two week wind drought, hydro could supply almost half energy demand although not peak power, even if there was little wind and solar.  California gets less than 10% of its electrical energy from wind so there is more than enough hydro to back up all its wind

4. Most of California's variable renewable energy will come from solar and surprisingly enough the sun shines every day so the idea that you would need to cover two weeks of energy needs from storage is just fanciful. 

5. Voltage and frequency from inverter based systems is far more stable than rotating systems and by oversizing turbine rotors and similarly using high DC/AC ratios on solar farms output fluctuations can be reduced. The loss of a 1,000 MW coal or Nuclear plant or 750 MW CC gas plant will cause a far sharper shock load on the grid than any cloud or drop in wind can so the claim that renewables cause fluctuations in grid voltage and frequency is just wrong.

6. Because a gas turbine takes about 5 minutes to go from 30% load to 90% and a battery can go from charging at 100% to discharging at 100% in about two tenths of a second.  Experiments in Ireland and South Australia have shown that a battery of a given power rating can provide the same grid stability as a gas turbine 10 times its size. In other words this 300 MW battery will provide as much short term frequency/voltage control as 3,000 MW of spinning gas turbines. 

7. The first stage of battery installations are to serve two purposes

a) every day smooth the output of both renewables and FF plants so for example OC gas plants and particularly CC gas plants can run more efficiently

b) replace closed FF plants for peak supply on 2-5 days per year. During California's recent rolling blackouts it was short a peak of about 4,400 MW, so this battery alone would replace about 8% of the shortage. In the meantime new wind and solar plants, particularly tracking and east west fixed solar will continue to push peak demand down. Many of these plants will have their own storage in addition to the large plants being installed by the utilities.

In summary you are correct about one thing "we are being lied to about renewables" by people who don't understand the difference between MWh and MW, namely you

Nathan Wilson's picture
Nathan Wilson on Jan 26, 2021

"So my rough rule of thumb is that you need storage of 7 days times peak need."

Ok, so that's obviously an absurd straw man to be attacked to prove a point.  Such an over-simplified argument obscures the real issues we should be discussing.

Big questions like energy system decarbonization require careful study, and many appropriately careful studies have been done.  The latest is from Princeton University (here), and is discussed on this blog here.

You're absolutely correct that 7days of batteries is the wrong answer.  The Princeton study found that 5-8 hours of storage (scenario dependent) was enough.  But you need backup from "thermal" power plants ("Wind, Water, and Sun" with massive pumped hydro backup has been widely discredited).  These thermal plants can be run on hydrogen, fossil fuel with CC&S, or nuclear, with the winner depending strongly on assumptions.  A system relying only on renewables and hydrogen however, by far uses the most renewables, the most land, and has the highest environmental impact.

As for system cost, the Princeton study, as does most others, projects that the decarbonized energy system will cost more than the business as usual/Ref scenario.  That includes extrapolation of falling costs of renewables and batteries.

So obviously, wealthy places like California and Germany can choose to decarbonize their grids, but the problem is the rest of the world.  Outside of the clean energy community, many observers will say the largest problem facing people of the world is global poverty. 

The Green Jobs mantra is just a scam.  If it takes more jobs (people) to generate that same result, that will increase poverty.  However there is an important difference between creating foreign jobs (imports) and domestic jobs.  We won't decarbonize the world by importing Chinese solar panels.

The other issue for the developing nations is that they tend to have weak regulatory enforcement.  That means that CC&S is hopeless; rather than pay to sequesters CO2, companies will simply dump it (it is invisible and untraceable).  There is also a concern about environmentally friendly disposal of waste chemicals from solar panel manufacturing and battery disposal/recycling (if it costs extra, it won't be done).

Clean energy in general has the problem that it trades future fuel cost for up-front capital cost.  For developing countries, it's usually better to defer expensed into the future, not pay them up front.

The result is that many developing countries (e.g. India) are building a few high-profile solar PV projects, and massively expanding their coal industries to keep the lights on.

We could best help the world by not only cleaning up our own energy system, but by also exporting frac'ing technology, since the US experience is that fossil gas from frac'ing is cost competitive with coal for levelized cost, much lower in upfront cost, but enormously cleaner (for air pollution and health impacts) and somewhat lower in CO2 emissions.  We must preserve our existing fleet of nuclear plants; maintaining them is quite cost effective compared to other clean energy options (claims that they are not economical are really from the perspective of competitors who want to reduce over-supply in order to boost wholesale prices).  We should also be massively building and exporting/licensing Gen III+ nuclear plants, as these can be built with domestic labor and mostly domestic supply chains in many countries; also their safety depends only on the laws of physics, not strict regulatory enforcement.  And there is plenty potential for steep cost reductions in future nuclear technology; current plants use far less land, concrete, and steel compared to the renewables that are now beating them on cost.

Priceton study summary

Bob Meinetz's picture
Bob Meinetz on Jan 25, 2021

Thanks Nathan, all good points.

"The Green Jobs mantra is just a scam.  If it takes more jobs (people) to generate that same result, that will increase poverty."

And it's not just in foreign countries. For a moment let's consider the social equity implications of renewable energy in the U.S.

Renewables advocates here tend to support making electricity more expensive when it's most useful. They even consider these programs (labeled demand-response) as being equivalent to energy - as if making energy less available is somehow making more of it. As if.

When rich people (and I would wager most people reading these words, by global standards, are filthy rich) get a higher electricity bill, they shrug their shoulders, curse their utility, and pay the bill. When poor people get a higher electricity bill, it's a problem. It means both Mom and Dad have to work an extra hour each week to be able to do the laundry, or use their electric oven, before 10PM.

Renewables increase dependence on natural gas. Who lives near gas plants, and will have to breathe more SO2, NOX, and other harmful emissions coming from their smokestacks? Not the people living in Napa Valley, but those living in Signal Hill - predominantly, low-income people of color.

Then there's cross-subsidization. Who has solar panels on their roof? Not couples living in apartments in East L.A., that's for sure. They're forced to pay higher taxes to finance the lavish subsidies bestowed upon couples in Santa Monica who install them on their homes. After installation, those couples get electricity for free, and pay nothing to maintain the grid used by both groups.

Renewable energy: energy created for/used by rich people, at the expense of everyone else.

Nathan Wilson's picture
Nathan Wilson on Jan 26, 2021

Demand-response: that's a huge uncertainty in the decarbonization studies.  Princeton seems to believe people will tolerate grid control of BEV charging (maybe/partly) and water heaters (unlikely, I think).  Fixed scheduling (such as would be adequate in a nuclear-rich grid) seems fine for BEVs, but real-time grid control of millions of distributed loads will definitely make the grid more vulnerable to cyber attack.

Roof-top solar?  That's a huge policy failure.  Utility regulators are literally making the most expensive type of solar competitive, in the name of supporting renewables and green jobs.  I suppose southern California is finally revising rate plans to reflect that PV system generate power at time of the day when it is least valuable (source); but there is still an enormous false incentive. Home PV systems make electricity that is worth 22¢//kWh or more to the homeowner; but it displaces utility grid power that only costs 5¢/kWh or so to generate.

Peter Farley's picture
Peter Farley on Jan 26, 2021

1. Most grids around the world have had demand response programs for many years and they work. They reduce the investment in rarely used assets on the grid, reduce iron losses in oversized transformers and efficiency losses in oversized generators. You are partly correct it is uncertain how far demand response can be extended, but it is absolutely clear from at least 50 years experience that they do work.

2. Home PV systems in the US cost 22c/kWh because of arcane regulation and resistance from utilities. Rooftop solar in Germany and Australia come in at about US 10c/kWh, which is less than the cost of transmission, distribution and retailing/metering. Feed in tariffs that are much lower than retail prices encourage east west solar alignment, customer load management etc. so there is no need for a false incentive. In fact rural electrification and subsidised grid power for farmers and villages in places like India and Africa actually adds to the cost of energy for the whole community. Your 5c energy at the turbine is less than half the cost and often less than 1/3rd of the cost at the customer meter, so again a false equivalence 

Nathan Wilson's picture
Nathan Wilson on Jan 26, 2021

Certainly rural electrification is the most promising area for home PV systems.   A couple of miles of power line can easily cost more than a house, so these really need to be shared by neighborhoods, not individual homes.

But no, 5¢/kWh at the turbine vs. 22¢/kWh at the home is not a false equivalency for marginal load.  The difference is entirely costs that don't go away when the load changes.  That expensive distribution network is absolutely needed, regardless of whether the capacity factor is 1% or 100%.   The same goes for the billing system, the fleet of repair crews that keep the system running, the backup generators, etc.

Historically, billing plans have tried to encourage conservation by allocating fixed costs mostly to heavy (average) users.  But in a world of home PV, we have no choice except full truthfulness in billing.  Demand-response drives us in the same direction; if wholesale prices hit zero or go negative, what sense does it make to charge retail customers 22¢/kWh?  The fixed costs of grid access are dominant, and the energy cost is trivial, especially at times that the total system load is well below peak.    That's where all utilities will head.

Peter Farley's picture
Peter Farley on Jan 28, 2021

I agree with much of what you have said but if rooftop solar and batteries anywhere in the system but particularly near the load can be used to reduce peak demand, then the investment in generation, transmission and distribution can be reduced, thus reducing fixed charges to all consumers. 

For example in Australia a 300 MW/600 MWh battery is being installed in Victoria. That will earn money every hour of the year in FCAS services and every day in arbitrage and then a few days a year will operate flat out for 30-60 minutes and at declining load for another 2-3 hours. A peaker gas plant in the same location, based on current gas plant utilisation might get 10-12% utilisation and falling as home solar, utility wind and solar increasingly squeeze gas out of the day to day market. Thus the cost to provide 300 MW of peak day generation margin from the battery is a little over half the price that would be required to induce private investment in a peaking gas plant.

 

Bob Meinetz's picture
Bob Meinetz on Jan 29, 2021

"Most grids around the world have had demand response programs for many years and they work."

They work by transferring responsibility for providing an adequate supply of electricity, at any time of the day or night, to customers as a cost in inconvenience. Might not matter to you, but it matters a lot to lower-income customers.

"Rooftop solar in Germany and Australia come in at about US 10c/kWh."

Then why does Germany have the most expensive electricity of any non-island nation in the world?

"In fact rural electrification and subsidised grid power for farmers and villages in places like India and Africa actually adds to the cost of energy for the whole community."

But then, electricity that's available any time of day or night, in any weather, is worth a lot more than intermittent, unreliable, unpredictable electricity. Isn't it?

Peter Farley's picture
Peter Farley on Feb 11, 2021

1. They work for the same reason that offpeak power prices work they spread the load and reduce investment. By reducing excessive investment, they actually improve system energy efficiency and reduce investment levels thereby reducing fixed costs to every consumer.

Do you build roads so big that there is no increase in travel times on the busiest days, of course not so with any sense you change the incentives in the electricity system to reduce peak demand.

2. Germany has always had high retail power prices so they could a) encourage energy efficiency and minimise the cost of setting up their energy system and b) give big discounts to their export industries and c) they added all the high early costs of green schemes to retail customers only. Lrage industrial users skipped out. However in response to high energy prices German buildings and appliances are much more efficient than American ones, so their average annual expenditure on electricity is significantly lower share of their income than for the average US consumer

3. I am not suggesting people should be denied power I am trying to say poor countries have many demands on their resources and in most cases better schools and health services and a limited energy service are better value than a gold plated western style grid. Even in Australia now hundreds of farms are being taken off grid and most of those farms would have more electrical demand than villages of hundreds of people in poor countries.

Most poor villagers in India and Africa cannot afford grid supplied electricity, even though they generally pay half to 1/3rd the actual delivered cost. It is far cheaper to provide solar and small batteries and in some cases micro wind or hydro or a methan digester so they can charge some phones run a few lights and a television for 4-5 hours after dark and later possibly a small fridge. As economic development increases and customers can afford more energy, a grid connection may make sense

Peter Farley's picture
Peter Farley on Jan 26, 2021

All nuclear advocates claim there is enormous potential for cost reduction but it is rarely realised. In the meantime wind, solar and storage keep getting cheaper. So for the cost of a 2.2GW 17,000 GWh/year plant Vogtle you can build 7 GW of distributed wind 7GW of distributed solar and 5 GW/24 hours of pumped storage. This will supply 38,000 GWh per year and minimum available summer output of twice that of the nuclear plant.

It uses no water, the nuclear plant including the mines and nuclear fuel plants use as much water as 1,000,000 people.  The renewables need more materials at the start, about double that of the nuclear plant, but the nuclear plant requires the mining of about 5,000 tonnes of uranium ore per year so within 10 years the material cost equation comes out about even.

India is not massively expanding coal generation. many of the new coal plants are stalled and utilisation of operating plants is falling to around 55% from a high of 73%. In the first half of last year India closed more plants than it opened. Meanwhile it has 49 GW of wind and solar under construction. The reason India is doing this is that wind and solar tenders are coming in at less than the operating costs of coal and gas plants. That is why 40 GW of Indian coal plants cannot repay their lenders, because their margins are so small    

 Surprisingly the renewable plants also need less land. Plant Vogtle is about 5 square miles + its share of the nuclear enrichment plant + the mine. A 4 MW class wind turbine and access tracks uses about 3,000 square feet. within that space you can also put two 40' containers with batteries. About 1,000 such wind turbines will supply the same amount of energy as Plant Vogtle 3 & 4. and use 3,000,000 square feet or 1/10th of a square mile.  According to NREL 14% of US roofs covered with solar would supply 30% of US electricity demand with zero net land use.

It is true that old poorly designed solar feed in tariffs mainly benefitted the wealthy but in Australia the main buyers off rooftop solar have been middle to low income homeowners who can see that a 5-7 year payback on rooftop solar is a far better return than money sitting in a bank account or even added to the mortgage still saves them money. Even for very low income households, now it turns out that poor renters can pay enough extra rent to give the landlord a return on the investment in solar while reducing utility bills by almost double the rent increase. 

You are correct that most existing nuclear plants should be maintained as long as safely possible, but new nuclear is impossibly expensive in almost every case. This is obvious in the UK where Hinckley point has a 35 years contract at Pds92/MWh + inflation backdated to 2012, whereas Dogger bank wind farm is earning Pds 48 for only 15 years and then it has to take market price. Over the life of both projects nuclear power will be two and a half times the cost of wind power

Nathan Wilson's picture
Nathan Wilson on Jan 26, 2021

Sure, there are many examples of low renewable prices.  But somehow, zero examples of full scale grid decarbonization using renewables.  Even when component costs are low, the total system cost comes out high.  Variable renewables have very high external costs that advocates ignore, but utilities can't.

Bob Meinetz's picture
Bob Meinetz on Jan 28, 2021

"All nuclear advocates claim there is enormous potential for cost reduction but it is rarely realised. In the meantime wind, solar and storage keep getting cheaper...This will supply 38,000 GWh per year and minimum available summer output of twice that of the nuclear plant.

Comparing dispatchable energy to intermittent energy is apples/oranges. Throw in the cost of the natural gas to back them up, and we have a comparison.

Do you use electricity only when the sun shines or the wind is blowing? (Didn't think so).

"It uses no water, the nuclear plant including the mines and nuclear fuel plants use as much water as 1,000,000 people."

Nonsense. Every drop of saltwater used at Diablo Canyon is returned to the sea.

"Surprisingly the renewable plants also need less land."

Ridiculous. You continue to make these specious claims with no basis whatsoever.

Peter Farley's picture
Peter Farley on Feb 3, 2021

Comparing energy sources which have their own distinct profiles is quite relevant. Nuclear plants are very expensive and don't follow loads very well, even if their marginal cost of production is low, therefore an optimal grid will have other forms of generation and or storage to a) back them up during outages and peak demand and b) store excess electricity when demand is low and c) provide rapid response in the case of a large swing or trip in demand or supply.

Wind and solar are cheap and have very low marginal operation costs but they are intermittent, not unreliable-technical availability on wind and solar plants is usually higher than thermal plants. But because they are intermittent they also need backup.

However some of the backup is already being built into the system, because wind and solar conditions are never the same over a whole state let alone a whole country and wind is stronger at night and in winter and solar the reverse, so the combination of wind and solar is never zero.

Further new solar panels are so cheap that it is often economical to add more panels than the grid connection is rated for, so even if some cloud passes over, or at 4pm on a summers afternoon a tracking solar plant still delivers its rated power. Similarly modern wind turbines cut in at about half the ground wind speed that the old design units did because they are taller and have larger more efficient rotors. For the same reason they reach full power at much lower wind speeds so a modern 2MW wind turbine generates 40% more power over 50-60% more active hours than a ten year old version.

As wind and solar capacity climbs, more and more often plant output will be limited below available capacity just as it is today with hydro, gas, coal and even nuclear plants. That means the wind and solar plants will contribute to synthetic inertia and primary reserve even without added storage. So looking at the eastern Australian grid the worst wind and solar week was about 60% of the average wind and solar week. As modern technology introduced and the geographic diversity increases, that number will rise for the reasons stated above. In fact you can see it in the data in Australia, Britain, Spain and Germany that worst renewable weeks in recent years are getting closer to average renewable weeks.  To top all of that of dispatchable renewables like hydro, geothermal, biomass and waste to energy can be turned down during high wind solar periods and turned up higher than average when wind and solar are low, so even if the combined output of dispatchable renewables in only 10% of supply over a year it can be 20% for a week, thus significantly reducing the need for new storage or other forms of backup.   

Peter Farley's picture
Peter Farley on Feb 11, 2021

I repeat, 1,000 5 MW class wind turbines each with three 40' containers of batteries take up about 3,000 square feet each. i.e a total of less than 70 acres. They generate the same annual energy and have a higher summer peak power rating 3-6 GW than plant Vogtle III & IV. 2.2 GW. Plant Vogtle uses 3.3 square miles, or 2,100 acres. 

Mariano Equizzi's picture
Mariano Equizzi on Jan 30, 2021

In two minutes eager reading a picture nobody wants to see and only for corruption.

Roger Arnold's picture
Roger Arnold on Feb 1, 2021

Replying here at the outer level, since nesting has gotten too deep.

One of the unfortunate features of Dr. Wojick's article is that he overstates things like the cost of grid-scale battery energy storage when his case would be valid even had he understated everything. The exaggerations in the article are distractions. They open the door to endless quibbling about points that don't matter to his basic point -- which is (or should be) that it simply isn't economically feasible to get to a 100% carbon neutral energy system through any buildout of wind, solar, and conventional battery storage.

It doesn't matter whether the cost of grid-scale battery storage is $1500 per kWh or $500 per kWh. It doesn't matter whether the calculation for storage requirements are based on supplying full peak load for X hours or only 1/2 of peak load. It doesn't matter whether the analysis includes contributions from hydroelectric or geothermal. They are far too small to affect the overall conclusion. A factor of 2 here and there is irrelevant, when the feasibility shortfall is a factor of 100 or more.

Let's be clear: grid-scale battery storage, even at today's prices, does deliver value. It enables wind and solar to contribute a larger fraction to supply than would otherwise be possible without extensive overbuilding and curtailment. It reduces the system's net carbon emissions per MWh. (OK, I know that's an arguable statement under full life cycle accounting systems, but let's not get into that.) What it does not do is enable that contribution to reach 100%.

The "storage" part of "renewables + storage" is good for reducing the frequency and speed of cycling required from the fossil-fueled power plants that remain on the grid. Those plants are still required to meet demand when wind and solar are non-producing or under-producing. In many cases, storage can allow relatively efficient combined cycle gas turbine plants to deliver a larger fraction of overall energy supply, while inefficient peaker plants deliver less. It reduces net carbon emissions. But it does not reduce the net cost of electricity. It only reduces the duty cycle of the fossil-fueled resources that have to remain available to meet demand when wind and solar are not producing.

A good MIT study on the amount and cost of energy storage that would be required for a 100% renewable energy economy was published in 2017 in the scientific journal Joule. A report on that study was published in IEEE Spectrum -- link here. The cost required for feasibility: $20 (or less) per kWh of energy storage capacity.

Bob Meinetz's picture
Bob Meinetz on Feb 1, 2021

"A factor of 2 here and there is irrelevant, when the feasibility shortfall is a factor of 100 or more."

For some, whether it's a factor of 2, or 100, or 1,000 doesn't matter. They say, "We'll wait for the price to come down. It's just a matter of time." To which I say, " I won't wait, for the very same reason."

Peter Farley's picture
Peter Farley on Feb 3, 2021

 I do agree that storage makes the whole grid work better as can be seen by the dramatic increase in reliability and fall in costs in South Australia and as you say it will keep CC gas plants on line longer 

The objection to the rest of the thesis is it is not a factor of two or even ten.

1. Tesla is saying they are quoting $200,000 to $300,000 per MWh now and expect to reduce battery costs by 50% over the next three years, Zinc, Aluminium and Sodium are all at some stage of development and will all be cheaper than lithium batteries so $120,000/MWh is a reasonable target within 5 years

2. Because of the overlap of wind solar and hydro and the inevitable excess capacity in behind the meter solar and batteries that customers buy for all sorts of reasons . together with the rapidly declining cost of flexible demand, the author has overstated the quantity of battery storage not by a factor of two but at least twenty. 

In combination the author has overstated the investment required in grid batteries by a factor of at least 200.

3. Studies by Rocky Mountain Institute among others have shown that a renewable grid with similar overcapacity as the current thermal grid has can supply 80-90% of demand without new storage so even if 10% of the energy passes via storage and that storage triples the cost of the power it delivers the storage has added 20% to the cost of power.

However as wind and solar power are now being contracted at the between 30 and 80% of the cost of new coal and gas generation and a quarter of the cost of new nuclear, (let's say 60% of the average conventional mix) then the cost of power will be 120% x 60% =  70-75% of a replacement conventional grid 

 

Roger Arnold's picture
Roger Arnold on Feb 3, 2021

$200,000 to $300,000 per MWh of energy storage capacity from Tesla is consistent with what I've read as well. Specifically, $200,000 for the battery modules alone, and $300,000 when the balance of system (mainly power electronics) are included. I agree that $120,000 per MWh in five years is a credible estimate for large battery storage systems in 5 years.

It's not clear whether that will be from some variation of Li-ion or from one of the new candidate technologies, but I expect that a commercially viable technology at that price point will emerge within 5 years. However ..

A study from a team at MIT was published in the scientific journal Joule in 2017, regarding the cost of storage required to make stored energy economically competitive with fossil-fueled energy for backing renewables. (I.e., to make 100% renewables economically competitive). The cost figure that they derived was $20 / kWh ($20,000 per MWh).  If they're right, then $120,000 / MWh, is still a factor of 6 too high.

I haven't reviewed the MIT paper myself. A report on the study was published in IEEE Spectrum about 2 years ago. The title was "How Inexpensive Must Energy Storage Be for Utilities to Switch to 100 Percent Renewables?"

I haven't reviewed the RMI study that you mention, but in general, I tend to be skeptical of anything they publish. I don't consider them objective.

Bob Meinetz's picture
Bob Meinetz on Feb 5, 2021

"A study from a team at MIT was published in the scientific journal Joule in 2017, regarding the cost of storage required to make stored energy economically competitive with fossil-fueled energy for backing renewables. (I.e., to make 100% renewables economically competitive)."

Roger, replacing fossil fuels on an area-wide grid with a) intermittent unpredictable energy, or b) the same energy backed up by batteries charged by intermittent, unpredictable energy will never happen.

To anyone who disagrees, I offer this challenge: show me one country, one state, one community, one village, or even one home that runs on exclusively renewables + batteries. The renewables and batteries could be free - doesn't matter. There is no substitute for reliable, predictable energy that can be dispatched on demand.

Renewables activists, enraptured by the idea batteries at a solar + storage" project are being charged exclusively by its solar panels, point to projects like the following:

"Last week, the city of Los Angeles inked a deal for a solar-plus-storage system at a record-low price. The 400-MW Eland solar power project will be capable of storing 1,200 megawatt-hours of energy in lithium-ion batteries to meet demand at night. The project is a part of the city's climate commitment to reach 100 percent renewable energy by 2045."

There's just one problem: they aren't. The batteries are being charged by a grid mix, and are only positioned next to a solar farm to look "green". That Eland I & II, the worlds largest solar + storage facility, in one of the sunniest areas on Earth, can't provide a reliable supply of electricity without relying on fossil fuel backup should be proof - solar and wind will only extend the use of fossil fuel, and doom any effort to rein in climate change.

john Liebendorfer's picture
john Liebendorfer on Feb 6, 2021

Bob you asked for   "show me one country, one state, one community, one village, or even one home that runs on exclusively renewables + batteries.”     

At this time in the transition to renewables we don’t need to or is it possible to eliminate all fossil fuels.  However there are many examples of getting almost there.  And here are some.

Tasmania Australia.     https://reneweconomy.com.au/tasmania-declares-itself-100-per-cent-powered-by-renewable-electricity-25119/  And yes Hydro is a renewable energy

There are thousands of homes that exit off the gird with renewables. Just google 'off grid housing' for examples - here are some links    https://www.dwell.com/article/off-grid-sustainable-homes-d5e71ce5  

Also there is Passive House movement with tens of thousands of house around the world that are net energy zero. https://en.wikipedia.org/wiki/Passive_house

Yes some of these houses have backup generators.  But they are producing 95% of their power on site.  You are correct we are decades away form complete elimination of fossil fuels. But you seem to argue we have exploited all opportunities and are at the end of possibilities, rather than at the beginning of the transition. 

Bob Meinetz's picture
Bob Meinetz on Feb 9, 2021

"At this time in the transition to renewables we don’t need to or is it possible to eliminate all fossil fuels. However, there are many examples of getting almost there."

John, do you see that galvanized steel chimney sticking out of the very first of "Stellar Homes That Venture Off-Grid" at your link? That's attached to either A) a natural gas generator or (more likely) a B) 500-gallon tank of propane on the other side of the house. It allows its owner to fire up a stove or generator whenever he needs it.

And no, he is not "almost there". He's not "producing 95% of his power on site." Absent any evidence to the contrary, I would bet my bottom dollar he's generating even more carbon emissions than the same house would if connected to the grid.

Renewable energy is a lie, and renewables + batteries? That's a damn lie.

john Liebendorfer's picture
john Liebendorfer on Feb 9, 2021

I encourage you to take a second look.  That steel chimney you see clearly goes to a fire place in the living room.  As do all, but one, of the chimneys in seven of the nine house. Only one has what appears to go to a gas heater.  Two have no chimneys at all.  All of those fireplaces burn a renewable bio fuel.

Your confidence that it's not possible to reach 95% self sufficiency is without any factual support.  I know of several, both new and retrofit homes near me that achieve that.

I have question for you.  Can you name one country, city, state or providence anywhere that runs on 100% nuclear?  

To follow up on 100% renewables  Here is a list of 36 Countries, states or cities running on all or near all renewables (scroll down to bottom of article for list. ) 

https://en.wikipedia.org/wiki/100%25_renewable_energy#:~:text=The%20first%20country%20to%20propose%20100%25%20renewable%20energy,from%20hydroelectricity%2C%20Iceland%2072%25%20hydro%20and%2028%25%20geothermal%29.

Bob Meinetz's picture
Bob Meinetz on Feb 10, 2021

"All of those fireplaces burn a renewable bio fuel."

John, I take it by "renewable bio-fuel" you're not referring to sticks and leaves, but old-growth timber. Currently we're chopping down trees faster than they could possibly be replaced -  5 billion trees/year in the U.S. alone. No, trees are not "renewable bio-fuel" by any stretch of the imagination.

"Your confidence that it's not possible to reach 95% self sufficiency is without any factual support."

Your request I prove a negative is the hallmark of a bankrupt argument. Can you prove I can't fly by flapping my arms? Of course not, but I wouldn't ask you to try (I can't).

"I know of several, both new and retrofit homes near me that achieve that."

I've encountered many people who know of such examples - but somehow they're always unable to provide evidence to support their "knowledge".

Can you name one country, city, state or providence anywhere that runs on 100% nuclear?  

There aren't any. However, in the 1970s-80s France, Sweden, and Belgium achieved >70% carbon-free electri ity in less than 20 years - faster than any country in the world, before or since. They did it with nuclear energy.

"To follow up on 100% renewables:  Here is a list of 36 Countries, states or cities running on all or near all renewables..."

Because renewables/natural gas/anti-nuclear activists have deluged Wikipedia with propaganda, sometimes it's necessary to dig a little deeper to get accurate information.

As an example, I chose Tajikstan, the most populous example on your list, with a population of 8.7 million. The International Energy Agency, a source with a bit more credibility than Mr. Anonymous Wikipedia Editor, provides a graph (below) showing the country doesn't get all its energy from renewable sources, or nearly all its energy from renewable sources. It gets more than half of its energy by burning coal and oil:

What's more, the light green band of hydro is virtually the same width it was in the year 2000. That's because hydro isn't scalable - there's a finite amount of energy you can count on from natural waterways. Because of that unfortunate fact, since 2009 the country has become increasingly reliant on coal.

Re: the other references, I don't have enough time to verify them, nor would I bother. It's as waste of time for me or anyone else.

Peter Farley's picture
Peter Farley on Feb 5, 2021

Roger 

I agree with much of what you say but just as you view RMI with a little scepticism, having been a lifelong admirer of MIT I have been amazed at how they seem to have been woefully wrong on the energy transition. I don't know why such an elite institution has such a blindspot or maybe I only see much of their pro nuclear studies.

May I also say that if the final optimum energy grid has 5% natural gas and 10-15% nuclear I would not be opposed to that. Once we get to that point it would be far better to spend money on decarbonising other aspects of the economy than chase down the last few emissions from power generation.

Returning to the MIT study, like you I couldn't find the original. I think at that time the full impact of rooftop overcapacity in solar and wind wasn't understood. For example in Australia the average household uses 3.8 MWh per year but the average solar system being installed now, produces about 8-9.5 MWh, maybe there is some bias in the samples but not that much. Some of the excess is exported but more and more exports are being limited by arbitrary caps. Gradually people start to learn to precool, preheat their houses, change the timer on their water heaters, dishwashers and pool pumps if they have them, charge their EVs etc so their energy imports fall lower and lower.

Similarly commercial wind and solar farms can configure their systems to provide FCAS services such as output regulation up and down, reactive power, synthetic inertia etc so they can make money even if not supplying energy. In summary, it is quite likely that in the equilibrium system, 20% of wind and solar will be curtailed just as roughly 40-50% of potential coal and gas power is curtailed now

Similarly businesses and institutions install batteries to mitigate demand charges and provide short term backup power. Transmission companies install storage to act as stabilisers on transmission lines, increasing their annual capacity. Distribution utilities install batteries to avoid upgrading transformers and feeders and act as rooftop solar sponges. For all these uses, energy storage of renewables is a side benefit but it is still available when needed. 

On the macro scale a detailed study by the US hydro industry and the Energy Department concluded that annual production could be increased by 10% and peak power by 30% with no new dams.

Small batteries eg 10-20% of peak power 30-60 minute duration at hydro plants allow them to participate in FCAS markets without using water and increase their power revenue without increasing transmission investments. Floating solar or near shore solar on northern and eastern shorelines can double the annual output of a hydro system with no investment in new transmission and little risk of congestion so that is another huge opportunity for "despatchable" renewables with little or no new storage.

Also by installation of new runners and flow controls and upgrading the generators some hydro systems have been able to upgrade the peak power of existing generators by up to 30% while reducing the minimum stable operating output. A few years ago operators of the Columbia River hydro system were curtailing wind because they couldn't safely curtail hydro sufficiently. With smarter hydro that problem is reduced so there is more water when wind and solar are low.

In conclusion, if the MIT study was run now allowing for excess wind and solar, behind the meter batteries and more flexible hydro it would almost certainly lower the cost of remaining storage, probably by a factor of three to five.

 

 

Bob Meinetz's picture
Bob Meinetz on Feb 9, 2021

Peter, I think the MIT / Argonne study you're referring to is

The value of energy storage in decarbonizing the electricity sector

It's summed up here:

The $2.5 trillion reason we can’t rely on batteries to clean up the grid

by, I guess, someone who is just as "woefully wrong" and shares the same "blind spot":

"Not only is lithium-ion technology too expensive for this role, but limited battery life means it’s not well suited to filling gaps during the days, weeks, and even months when wind and solar generation flags.

"This problem is particularly acute in California, where both wind and solar fall off precipitously during the fall and winter months.

"This leads to a critical problem: when renewables reach high levels on the grid, you need far, far more wind and solar plants to crank out enough excess power during peak times to keep the grid operating through those long seasonal dips, says Jesse Jenkins, a coauthor of the study and an energy systems researcher. That, in turn, requires banks upon banks of batteries that can store it all away until it’s needed.

And that ends up being astronomically expensive."

What are the chances researchers at two of the most prestigious technological institutions in the world - MIT and Argonne National Laboratory - could be right, and you wrong?

 

Peter Farley's picture
Peter Farley on Feb 11, 2021

I did find the IEEE Spectrum article that Roger referred to which claimed that storage needed to drop to $5/kWh if all generation was to come from wind and solar. Three of the four regions they picked as representative of the US had little or no access to hydro, so in terms of backup that is not representative of the whole US grid. However the same article says

But those figures are only for scenarios in which solar and wind meet power demand 100 percent of the time. If other sources meet demand just 5 percent of the time, storage could work at a price tag of $150/kWh. Which technologies could hit that target?

Nobody seriously expects us to shut down all the existing hydro, biomass and geothermal plants which already supply 9% of US electricity so even in 2016 the $5 figure was a straw man. 

In 2016, the concept of significant overcapacity in renewables, particularly in rooftop solar had not been seriously considered. Yet it seems that it is perfectly fine to run a system with significant overcapacity in conventional resources. The coal fleet in the US has a best ever recorded capacity utilisation of 67%.  In spite of almost 200 unit retirements (1/3rd of 2010 capacity) it is now running at about 42%. Combined cycle gas plants at about 55% and other gas plants - OC, steam and IC at an average of 13%. So if sufficient wind and solar is constructed to generate double US annual demand and 1/5th of output is stored and 1/3rd curtailed then storage requirements fall by a factor of about 5. The study did not consider thermal storage which is 1/5th of the cost of batteries or smart charging of EVs which is virtually free. Nor from what I can see did it consider the high capacity factor of offshore wind, so in fact while I don't disagree with the maths of the study, it is a limited subset and advances in generation technology and load control have made it obsolete.

 As an indicator South Australia has no coal, no hydro and no nuclear, it is currently running at 70% renewables and total storage is equivalent to 6% of peak demand for one hour. It has recently reduced spinning reserve requirements so could reach 75% renewables in the next quarter with no new utility renewable installations.

It expects to double its installed storage power and triple the duration over the next 3-4 years by which time it will be net 90-95% renewable. Oh and its wholesale power prices are the lowest in Australia  <US $25/MWh

By the way

1. Coal generation in the US has fallen from around 2,000 TWh in 2008 to 765 TWh last year (IEA) so where the increasing reliance on coal comes from I don't know 

2. US forest biomass has increased 40% since the 1990's (US Forest service)

 

       

 

Pages

Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »