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

Ten research teams aim for long-duration storage at 5¢/kWh

Ten research teams aim for long-duration storage at 5¢/kWh

To get long-duration storage costs down to 5 cents per kilowatt-hour, research teams funded by ARPA-E are pursuing breakthroughs in flow batteries, hydrogen storage and other technologies—even thermovoltaics.

DECEMBER 30, 2019 WILLIAM DRISCOLL

Ten teams working to drive down the cost of long duration storage are competing in a way, using federal grant support to make enough progress to earn a follow-on grant for pilot-scale production. Projects include a sulfur flow battery for full-week backup capability, and a more efficient means of converting electricity to hydrogen and back again.

Each project aims toward a goal of 5 cents/kWh for storage that can last for days, under the DAYS program of the U.S. Department of Energy’s Advanced Research Projects Agency (ARPA-E).

Observations:

  1. Finally somebody is telling the truth about storage batteries – currently it is all being researched.
  2. Since 1859, when our car cranking battery, lead acid technology was placed in service, there was no real progress.
  3. Using exotics materials with a price tag to match offered little or no competition to the lead acid old, reliable battery.
  4. Talking about using lithium ion batteries for storing power plants capacity is of course irresponsible chattering.
  5. All the known technologies, working or not, failed to meet the real test – they all are being not cost effective.

https://pv-magazine-usa.com/2019/12/30/ten-research-teams-aim-for-long-duration-storage-at-5%c2%a2-kwh/?utm_source=pv+magazine+USA&utm_campaign=2b8097a481-RSS_EMAIL_CAMPAIGN&utm_medium=email&utm_term=0_80e0d17bb8-2b8097a481-159206113

Discussions

Bob Meinetz's picture
Bob Meinetz on Dec 31, 2019 4:57 pm GMT

Noam, the Energy Information Administration gives an average installed cost of $1,500/kWh for Li-ion, medium-duration battery storage (more for long duration storage).

To get to 5¢/kWh, price would have to come down by a factor of 30,000.

Can anyone in the renewables community count to ten?

Joe Deely's picture
Joe Deely on Dec 31, 2019 5:36 pm GMT

To get to 5¢/kWh, price would have to come down by a factor of 30,000.

So Bob are you saying that each storage system can only be used once? 

Bob Meinetz's picture
Bob Meinetz on Dec 31, 2019 5:45 pm GMT

No, I'm referring to capacity, which you seem to have no problem understanding with wind and solar, but have a blind spot with energy storage.

Once your batteries run out of electricity, either the grid goes down or you're forced to rely on fossil fuel backup - not good enough. Since powering CA for even one cloudy day would require batteries costing more than $1 trillion, affordable capacity isn't even in the ballpark - no matter how many times your magical batteries can be recharged. Not in California; even less so in other states.

Please - refuting these nonsensical dreams of yours is wasting everyone's time.

Joe Deely's picture
Joe Deely on Jan 1, 2020 8:21 pm GMT

No, I'm referring to capacity, which you seem to have no problem understanding with wind and solar, but have a blind spot with energy storage.

I'm not the one with lack of understanding here. Let's look at what you said.

  1. Average cost of installed capacity = $1,500/kWh
  2. To get to 5¢/kWh, price would have to come down by a factor of 30,000.

If, as you said the price dropped by a factor if 30,000 the new average cost of installed capacity would be

      $1,500/kWh / 30,000 = $0.05/kWh.

We would now divide by the number of lifetime cycles to get the cost per cycle. Since we are already at $0.05/kwh you are implying that number of cycles will only be 1.  

Note: the DOE assumed an installed cost of $150/kWh in order to come up with cycle cost of $0.05/kWh.  They assumed a lifetime of 5,000 cycles. 

Perhaps you can show us how your assumptions and math differs from DOE when you said - "price would have to come down by a factor of 30,000."

 

Some people think we are already at  $150/kWh.

Battery prices, which were above $1,100 per kilowatt-hour in 2010, have fallen 87% in real terms to $156/kWh in 2019. By 2023, average prices will be close to $100/kWh, according to the latest forecast from research company BloombergNEF (BNEF).

 

Bob Meinetz's picture
Bob Meinetz on Jan 2, 2020 7:35 am GMT

"We would now divide by the number of lifetime cycles to get the cost per cycle."

Actually we wouldn't, because the cost per cycle is irrelevant.

Not sure why you're having such a hard time with this: it has to do with how much clean backup energy can be stored at one time to provide a steady supply of electricity to the grid when the weather is cloudy and/or calm.

Both a) allowing the grid to go down, and b) powering the grid with fossil fuel gas are no longer options. If enough clean energy can't be stored to provide one day of power when solar and wind aren't available, renewables + batteries are worthless for shutting down gas and coal capacity, and thus are worthless for stopping (or even slowing) the progression of climate change. As they always have been.

Nathan Wilson's picture
Nathan Wilson on Jan 4, 2020 3:46 pm GMT

Hmm, Bob and Joe are both partly right. As Joe says, the cost added to each kWh of energy delivered from storage is the initial cost of the storage divided by the cycle life (actually somewhat higher due to interest and losses). 

But as Bob points out, even with storage, a PV dominated grid will still need a lot of fossil fuel backup. It does not take much PV before the net demand peak shifts to cloudy/winter days. And storage for cloudy days will be much less cost effective than evenings (following sunny days), because it will get fewer revenue generating discharge cycles per year.

The need for backup power plants greatly weakens the economics  of solar, particularly in places with cheap coal and little gas, like China and India.

Windpower also decreases the benefit of storage, due to wind's irregularity. As a result, studies by the US DOE NREL predict very low use a storage in hypothetical renewal dominated grids, even with very low priced storage.

Joe Deely's picture
Joe Deely on Jan 4, 2020 11:32 pm GMT

 As Joe says, the cost added to each kWh of energy delivered from storage is the initial cost of the storage divided by the cycle life (actually somewhat higher due to interest and losses). 

Yep.

Note that in DOE calclations with cost of $150 /kWh and 5,000 cycles the cycles cost ends up being $0.03/cycle.  The boost back up to $0.05 is from losses, maintenace and interest as you said.

As a result, studies by the US DOE NREL predict very low use a storage in hypothetical renewal dominated grids, even with very low priced storage.

Any links?

This NREL report - although primariy focused on peaking capacity - seems to indicate otherwise. See Fig 4 and below.

We demonstrate the opportunity for utility-scale battery storage to satisfy a substantial portion of U.S. peak capacity needs and thus expand beyond its current role in the relatively small ancillary services market. This analysis demonstrates roughly 28 GW of practical potential for 4-hour storage providing peaking capacity, assuming current grid conditions and demand patterns. This deployment could help decrease storage costs—and storage deployed primarily to provide peaking capacity can provide additional benefits, such as a sink for low- or zero-value PV generation during non-peak periods. This in turn can enable greater PV deployment, which then increases the potential of 4-hour storage. This effect can extend the practical potential for 4-hour storage to 50 GW or beyond nationally (assuming PV provides 10% of the nation’s electricity demand). Of course, there could be significant regional impacts, as the areas first to adopt 4-hour storage could saturate their potential before full national deployment is reached. However, the general effect should provide additional potential for cost reductions to increase the competitiveness of 6- or 8-hour storage.

The results show significant potential for energy storage to replace peaking capacity, and that this potential grows as a function of PV deployment

 

Bob Meinetz's picture
Bob Meinetz on Jan 5, 2020 12:48 am GMT

"This analysis demonstrates roughly 28 GW of practical potential for 4-hour storage providing peaking capacity,"

Peaking capacity only? Not remotely enough. To shut down gas and coal capacity and maintain "4-nines" (99.99%) reliability you would need weeks of storage - not just one day - and the cost not just for storage, but the added renewable capacity to charge it would be at least $10 trillion for the state of California alone.

Please - renewables + storage will never serve as a viable replacement for any kind of dispatchable electricity. Not nuclear, coal, nor gas.

It's somewhat amusing that renewables advocates toss around storage as a viable option, when all of California could be powered with 10 Gen-2 nuclear plants for a total of $200 billion - a fraction of the cost of solar + wind + storage - and it would last at least ten times as long. 

DOE:

"...additional modeling work is needed to accurately quantify the impact of LDES on wind and solar penetration at the regional level, and should include realistic handling of grid power flow constraints, network stability, contingency requirements, opportunity costs of curtailed energy, limits to load flexibility, and other parameters necessary to capture the full complexity of delivering power within a large electricity system."

In other words, DOE has no idea whether renewables + storage could replace coal and gas capacity on the grid. When DOE has "captured the full complexity of delivering power within a large electricity system", do let us know. Until then, we need to move forward with what works.

Bob Meinetz's picture
Bob Meinetz on Jan 5, 2020 4:09 pm GMT

"As Joe says, the cost added to each kWh of energy delivered from storage is the initial cost of the storage divided by the cycle life (actually somewhat higher due to interest and losses)."

Nathan, you're confusing battery discharge with capacity (marginal cost vs. capital cost).

The marginal cost of operating a battery bank is insignificant - cost of maintenance, operation, etc. But the capital cost - the full cost of batteries plus installation - needs to be paid off over the life of the batteries, typically a maximum of 10-12 years, whether every kWh of capacity is used or not. Whether one kWh is used, or not.

The full bank of batteries has to be available when needed - the cost of availability, which renewables activists would like us to believe doesn't matter. It matters very much to the battery manufacturer, who demands full payment for his batteries plus interest over their lifetime.

As I've pointed out in this post and others, the cost of batteries able to power the CAISO grid in California for one cloudy day approaches $1 trillion at today's cost, not some number from a renewables accolyte's fevered imagination. The mortgage on $1 trillion + 4% interest over 10 years comes out to $10.165 billion per month, not including the renewables capacity necessary to charge them. Add that to Californians' electricity bills however you like: solar + wind + batteries is not remotely affordable now, and never will be.

Joe Deely's picture
Joe Deely on Jan 5, 2020 11:34 pm GMT

Actually, using $150 /kWh and 620,000 MWh as an average daiy load on CAISO means that the storage would cost $93B.

Of course, by the time we get anywhere near that amount of storage the cost will have dropped to about $50 /kWh

Bob Meinetz's picture
Bob Meinetz on Jan 6, 2020 1:29 am GMT

Like I said, "$150/kWh" is the product of some renewables accolyte's fevered imagination. It costs ten times that now - like I said, very close to $1 trillion for one day. Actually.

"Of course, by the time we get anywhere near that amount of storage the cost will have dropped to about $50 /kWh."

Bed and rest, Joe.

Joe Deely's picture
Joe Deely on Jan 7, 2020 9:44 pm GMT

Bob,

Always fun when you put numbers out there.

APS (Arizona) - is installing 423 MWh of storage this year.

 Energy storage increases flexibility by soaking up excess solar energy produced during midday and deliveringit to customers later in the day when they need it most.

According to you they are paying $635M for this storage.

Next year they are implementing 1,212 MWh of storage = $1.8B - using your cost.

If we assume solar at $0.02/ kwH then each cycle of storage is costing $0.52 /kWh if we use your costs.

APS will then turn around and sell this electricity to their customers for $0.24 /kWh. The more they sell the more money they lose.

What a great business plan.  I wonder if anyone has mentioned this to the AZ PUC.

Bob Meinetz's picture
Bob Meinetz on Jan 8, 2020 4:34 am GMT

It's even more fun when you embarrass yourself with them, Joe.

First off - they aren't my numbers. I know when EIA's hard facts don't favor solar you must shoot the messenger, but the cost of storage is theirs. Ready, aim, fire...

Second, you've got that Ouija Board out again. No, we don't "assume" anything, because that's usually just Joe making stuff up. For example, I have no idea how you arrive at gibberish like "each cycle of storage is costing $.052/kWh," and I don't really care. It was probably pulled from the same place you get most of your statistics.

Joe Deely's picture
Joe Deely on Jan 9, 2020 3:55 am GMT

For example, I have no idea how you arrive at gibberish like "each cycle of storage is costing $.052/kWh,"

Arithmetic Bob - arithmetic. Explained by both me and Nathan above.

Although I said $0.52/kWh not $.052 /kWh as you state in your comment.

It would be worth your time figuring out how to do this calculation. That way, when you see crap from EIA - you can call it out.

So for example if someone says battery cost = $1,500 kWh and you know the base Tesla 3 model has 50 kWh and a base price of under $40K - you can say - somebody is lying.

Or when you see the Google/NV announcement today you can understand the costs quoted in article and perhaps call BS if they don't make sense.

That includes the Gemini Solar + Battery Storage Project, a 690 MW solar facility with 380 MW of batteries, projected to cost more than $1 billion. 

...

NV Energy also has a contract with 8minute Solar Energy LLC for the output of the Southern Bighorn Solar & Storage Center, a 300-MW solar project with 135 MW of battery storage, at an average price of $35/MWh.

 

Noam Mayraz's picture
Noam Mayraz on Jan 9, 2020 1:33 pm GMT

This is a generic reply to all of the above contributors’ comments benefits.  

From above, a true statement: "Nathan, you're confusing battery discharge with capacity (marginal cost vs. capital cost)."

  1. I read the $0.05 (5 cents) /kWh) as the cost of the stored electrical energy.  Today solar PV producers sign PPA’s at 3 cents / kWh – just pure logic.
  2. We do not have electrical energy storage batteries – it is all talk no substance - therefore the actual cost of such batteries is irrelevant.
  3. In Hornsdale Wind Farm in S. Australia Elon Musk sold a useless bill of goods (something passed off in a deception or fraud).  Those batteries have limited storage, useless.  The system worked once for 40 minutes.
  4. We cannot use renewables and store that power too.  Concentrated solar power (CSP) could have done that, but that technology is currently DOA (dead on arrival). 
  5. Renewable sites’ batteries, if and when available, will have to be charged off the grid – we cannot eat our cake and have it too.
  6. Storage batteries are a wet dream for lack of technology and to boot excessive cost if someone would attempt installation of any batteries.
  7. Current batteries are used for grid quality, power factor mitigation and frequency stabilization.  Not for energy storage as some erroneously believe.
Bob Meinetz's picture
Bob Meinetz on Jan 15, 2020 11:19 pm GMT

Noam, if the Hornsdale battery is now delivering 40 minutes of power to the grid, it's an elevenfold improvement over an event in 2017 when it folded like a cheap suit after three minutes of service when a coal plant went down:
Fortunately, it was rescued by another coal plant before it could completely destroy either the SA grid or the fragile expectations of renewables advocates.

Joe, you do understand that 8Minute Energy's grid-scale storage facilities might not be storing only the meager generation from solar plants, but a grid mix with whatever combination of fossil fuel generation is coming along for the ride - don't you?

As a matter of fact, I contacted 8minute's L.A. office with that very question: are your batteries storing only clean solar energy, or a grid mix? The tech who answered the phone said, "Good question, I don't know the answer. Can I have someone get back to you?

"Of course," I replied, thinking waiting for an answer would make as much sense as trying to power a grid with batteries. And it would have.

Noam Mayraz's picture
Noam Mayraz on Jan 14, 2020 6:29 pm GMT

Bob, your last comment, way down below, was great.

That political lie that renewable energy will be use and then stored, EAT OUR CAKE AND HAVE IT TOO.  Is obviously a baloney.  Thx again, Noam.

Joe Deely's picture
Joe Deely on Jan 19, 2020 9:12 pm GMT

Joe, you do understand that 8Minute Energy's grid-scale storage facilities might not be storing only the meager generation from solar plants, but a grid mix with whatever combination of fossil fuel generation is coming along for the ride - don't you?

My assumption is  the same as yours Bob - if this storage were to go live today then a large share of the fuel mix on the grid at the time of charging would be NG. Note: most charging would probably take place between 10am - 2pm when electricity is cheapest.

But as per usual you "don't skate to where the puck is going".  Plus, you talk about this project in isolation vs looking at the bigger picture.

Here is a simplified version of the bigger picture:

Total Generation in NV = 40,000 GWh/yr 

Avg Gen per hr = 40,000/8,760 = 4.566 GW = 4,566 MW 

Currently NV Energy doesn't currently have a lot of solar - only 1,020 MW. Below is list of these projects.

Here is the list of upcoming solar - one of which is the 8Minute Energy BigHorn solar/storage project you referenced.

So that brings us to about 3,700 MW of solar vs. 1,020 MW today.

Now let's consider some other factors.

NV Energy:

- has about 500 MW Geo-Thermal capacity

- share of Hoover Dam is 238 MW

- rooftop solar is growing by 100 MW year

- NV energy is part of CAISO EIM and CAISO solar exports are growing yearly in spring/fall months.

How much share will solar have between 10am-2pm on NV grid in 2023? 

Do you see where the puck is headed Bob?

Finally Bob - I am confused as to why you are concerned about how these batteries will be charged and why you went to the bother of contacting 8Minute energy. 

The cost of PPA to NV Energy is $34 MWh- which is equivalent to $0.034 kWh

But you said:

To get to 5¢/kWh, price would have to come down by a factor of 30,000.

So, are you saying that price has actually come down that much or are you saying that 8Minute Energy is selling something for WAY more than it costs?

They'll be broke before they return your call.

 

Bob Meinetz's picture
Bob Meinetz on Jan 20, 2020 4:34 pm GMT

Joe, I'm not at all concerned how these batteries will be charged (I know how they'll be charged). I'm concerned about the ongoing scam that 8Minute, together with:

Southern California Edison
City of Palo Alto
PG&E
Southern California Public Power Authority
Los Angeles Department of Water & Power
San Diego Gas & Electric

are perpetrating upon the public and the environment, with the help of unethical solar entrepreneurs. Maybe you don't know what an offtaker is (if you don't understand the difference between capital and marginal costs of solar farms, it's very possible) so I'll fill you in.

The entities listed above are offtakers for 8Minute Energy. That means they buy the "renewable attributes" of solar energy generated by 8Minute's farms and "apply" them to electricity generated by burning gas, by coal, by oil, firewood, tires, whatever garbage they have on-hand to burn and make steam - then sell it to CCAs as "100% renewable" electricity.

"100% renewable"? How can they get away with that, Joe? Does buying these invisible "attributes" magically pull the smoke from their burning tires from the air?

You might say, "8Minute, after selling the attributes, probably sells the actual electricity generated by their farms as '100% Burning Tire Electricity' - they take responsibility for the emissions." That would make sense. That would be fair. Right?

Wrong. 8Minute sells the electricity from their solar farms as "100% renewable" too. Its so-called "renewable attributes" are double-counted, and 3.6 tonnes of carbon emissions/MWh, for which their solar farms are directly responsible, disappear into thin air.

I'd appeal to you to justify these crimes but I know you can't. So I won't.
 

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 »