as a resident of the J in PJM, I am delighted that they are finally understanding reality. of course, based on my electricity bills, I fear it is a little while before the benefits will be felt
Nice piece - fully agreed that the pendulum has swung with potentially a significant cost to Customers. The extreme focus on renewables over the past 10 years has given us a very expensive generation portfolio with less than desired reliability. The true cost has been both masked with all of the tax credits (transferred to federal government) and hidden by not fully recovering the costs of now retired assets (i.e., coal plants). Now as utilities and market pivot back towards a balanced generation portfolio and address the significant increase in demand driven by data centers/AI, utilities/IPP's are paying extremely high costs given the concentrated gas turbine demand.
- The cost of a spare turbine we've been evaluating increased by over 50% over the past 9 months.
- A recently received bid for a greenfield gas turbine resource was $1B (~20%) over the prior estimate.
Another way to think of it is that the mis-guided generation investment over the past 10 years driving elevated customer rates is going to take 10 to 20 years to resolve.
Another factor that will likely influence generation cost going forward is an issue that is frequently mentioned in the mining industry. The most attractive and lowest ore bodies have already been mined resulting in future mines requiring higher prices to justify economically. I see the same happening with the siting of new generation resources, especially here in the southwest with our limited water resources.
- Gas turbines being evaluated with expensive air cooled condensers as opposed to wet cooling
- New nuclear siting challenged by the need to identify sites with significant amounts of water to support meaningful plant capacities (2 x AP1000 requires ~44K acre-ft/year).
- Existing transmission is at or close to capacity resulting in new resources needing to add very expensive transmission as well.
- Attractive wind turbine sites that are closer to customer load have already been utilized.
In summary, cost of new generation will continue to increase independent of the soaring turbine costs.
“Say what you want about data centers.” Okay, I will. I have no problems with the technical analysis in this post. But what are the social and political costs of increased electricity use? I see the growth of data centers and AI causing massive increases in electricity demand. Corporate management and shareholders who already hold much of the wealth in our country benefit. Working class people, both in blue and white collar jobs, get laid off due to automation fueled by AI. The middle class foots more of the bill for electricity due to the expansion of data centers and AI. The share of the wealth in our country continues to shift from the lower and middle classes to the wealthy elites. Social unrest increases and political instability ensues. I hope I am wrong.
"The middle class foots more of the bill for electricity due to the expansion of data centers and AI. "
This does not have to be the case. There are several approaches available to isolate AI demand from the current grid, including "islanding" suggested by President Trump.
I see an intensive focus on ensuring the data centers/AI do not increase the cost for other customer classes. A number of ways this is being implemented:
- DC/AI customers fund the development of any incremental generation resources
- DC/AI customer absorb a meaningful portion of common costs (operational, transmission, and regulatory) incrementally reducing the amount allocated to other customer classes.
- DC/AI customers are required to agree to long-term contracts (20 to 25 years) with significant minimum payments (80% to 90% of planned load) as well as provide significant collateral in case they go out of business.
- DC/AI customers payments are in excess of the actual revenue requirement (additional security). If the customer does not sign a follow-on contract, this balance benefits customers and ensures they do not have to fund stranded generation assets.
Just my perspective from having visibility into multiple on-going DC/AI discussions.
I hope what you are saying is implemented. For now some communities are being hard hit by increasing electricity bills. And then there is also the issue of massive water usage by these data centers.
In the building phase thry will consume water same as any construction...but in the run phase the water is in a closed loop cooling system. The idea that DCs "consume" vast amounts of water is nonsense. It was dreamt up as a talking point by groups aligned to Neville Roy Singham. Please do a bit of reading on the subject before repeating tropes that may not be true. You can use AI to do that
Thanks for your input. I admit to believing DCs in general are huge water consumers without doing much research into the issue. I therefore did a deeper dive into the subject and am impressed by how much closed looped cooling can save water compared to evaporative cooling methods. According to one source I read most of the DCs currently use evaporative cooling. I hope closed looped is the default for future DCs. More than one article I read still found drawbacks to closed looped systems. There evidently is still water loss due to leaks and other issues but the closed looped beats evaporative cooling by a large margin. Another issue is that DCs are huge electricity consumers. If the electricity is supplied by plants using a lot of water (ie steam turbines), then the water usage issue is being relocated, not being completely mitigated. This is a fascinating subject.
Data centers need to generate and pay for their own power, and not burden the grid or consumers.
It's a little off topic, but you're right - the wealth gap just keeps getting bigger - and it's very hard on the working class, especially the working poor.
I have virtually no concern about income or wealth inequality
High income individuals get their income by providing goods and services that people want. They do not “take” money from the poor. They provide an option that people prefer. All unforced financial transactions benefit both the buyer and the seller, or the transaction would not happen. If a cup of coffee costs $4 and I buy it, it is because I value the coffee more than I value the $4. The seller in this case values the $4 more than the coffee.
I have many friends who are incensed at the wealth of Jeff Bezos. I am not. I buy most of the things I use from my local grocery store, Target, Kohls. Home Depot, Ace Hardware, Etc. They are close, they have what I want on hand, and I don’t generally have to make a special trip to shop at those stores. I buy things I purchase infrequently, things I can find at a store 45 minutes away (1.5-hour round trip just for the driving) or things I don’t know where to buy from Amazon. I get the item for what I would have purchased at a local store, if one exists, and I get something more valuable that I cannot buy for any price for free. That item is more time. I can’t buy a longer life, but I can reduce the time I spend acquiring some of the things I need or want. Thank you Jeff.
I say I have “virtually no concern” for a reason. A wealthy person can only do a few things with their money. They can:
- Buy goods and services themselves. The money spent there goes into someone else’s pocket, his or her local economy, etc. Wealthy people shopping benefits those they buy from and hundreds of other downstream people including many blue-collar workers. Those workers include people who mine materials, provide logistics, make subcomponents and finished goods and install or maintain any number of objects in a wealthy person’s home. All good.
- They can invest the money (probably the bulk of what they do). Investment grows the economy, increases productivity, and makes us wealthier. All good.
- They can pay taxes. People on the left would argue this is always good.
- They can donate money.
- Take money out of productive circulation. That includes money in their wallets, money in a safe at home, or I believe money in nonproductive assets like gold. I don’t know how large this pot is, but it is probably relatively small.
- Use their wealth to buy favors with government officials. I do not like this one. This one exists because our government is into too many things, makes too many rules, and interferes with our lives too much. A much smaller government would reduce the benefit of buying government favors.
The fact that someone else is rich, does not mean that I am poor or worse off. Imagine that the poorest among us all had $1 million a year in 2026 purchasing power but the richest had $1 trillion. Would we complain about the $1 million?
Income inequality is often inaccurately reflected by income numbers presented by the press. In the book “The Myth of American Inequality”, by Phil Gramm, he points out that incomes for the wealthy don’t reflect the taxes they pay (the top 20% of income earners pay almost all the income taxes) or the transfer payments that the bottom earners make. If I have $25,000 a year but get $10,000 in transfer payments, I am really making $35,000 a year. The wealthy have higher levels of family labor force participation and often work more hours.
Our understanding of things such as minimum wage are also wrong. The majority of minimum wage earners are younger than 25 or retired. They are not a family of 4 trying to survive on a minimum wage job. Young earners need entry jobs to gain skills which help them over their economic life’s. Minimum wages laws make those skill hard to acquire.
If we want to help the poor, we need to do things such as:
- Eliminate the minimum wage - Get rid of licensing for most occupations - Allow school choice so we can improve education benefiting all Americans - Kill the idea of college for all. Something like 40% of people who enter college never graduate. That is time and money poorly spent. I am not opposed to post high-school education but a degree in gender studies is financially worthless. Getting skills as a welder may translate to much better long term financial success.
I am concerned about wealth inequality in the United States since the trend has been a reduction of wealth for the middle class and an increase for the upper classes. The standard of living has worsened for the lower and middle classes. This is bad for the country for a number of reasons. Poorer health outcomes for the uninsured, housing affordability crises, capture of the government by wealthy elites. I could go on. I’m not anti-capitalist. But I think we can have a robust safety net for people without taxing the wealthy so heavily that they are disincentivized to work hard and innovate. I’m not a fan of trickle down economics. If the lower and middle classes are given a decent wage then doesn’t that benefit the economy as much as the uber wealthy benefiting “people downstream,as you put it, when they buy goods and services? Henry Ford evidently paid his workers a good wage so they could afford to buy cars which ultimately benefited Ford Motor Company. That’s trickle up economics.
The shift to natural gas combined-cycle generation will quickly require the addition of natural gas pipeline capacity to provide firm gas deliveries to those powerplants and perhaps also the addition of natural gas underground storage capacity.
Very accurate statement - development of the Desert Southwest Pipeline is a key enabler to add much needed gas generation in the southwest, at a current cost estimate of $5.6B.
Yep and in New England specifically, Massachusetts the 1600 megawatts of “offshore” wind have laid a giant egg and it isn’t golden. All these states and ISO’s responding to the mid wit politicians who pushed the solar and wind agenda are now seeing (whether they will admit failure or not is another story) cause and effect of the failure of the policies they pushed. Rate payers in New England will need to hold onto their hats as the python digests the mess of a renewable rabbit these fools created. Of course at the moment they are saying mess, what mess? Move along. Folks nothing to see here. It will get worse for the rate payers before it gets better, but like anyone kicking a bad habit it starts with recognizing the habit is bad and needs to stop. We are getting close to that point here in NEISO. Great post EBB’s!
For years we were told that for human needs reliability was not a big deal. Now that the big media company machines need reliability, it sits center stage. It is interesting how it helps to set priorities.
The demand for storage is interesting. But it agrees with my earlier statements that batteries are no longer a renewable item, and now simply a power system accessory. In this capacity they work well with fossil fuel generators, absorbing power when loads are low to keep units at optimum heat rate, discharging when loads are high to supplement generation.
I wish people would stop referring to Storage POWER Ratings (MW). Power Ratings are the capability to discharge electricity IF THEY HAVE THE ENERGY AVAILABLE.
Although important POWER is secondary to ENERGY CAPACITY (MWh). This is what tells you the rate & duration of discharge. In other words, the ability to meet demand after Wind and Solar fail.
All the POWER (MW) Capacity is meaningless if you do not have sufficient ENERGY to supply the POWER.
You're right, of course, as far as storage being capable of helping in any long duration wind and/or solar drought.
However, I'm working on a piece about the increasing use of storage as a short term (couple hours) daily energy leveler. It's feasible now, at least for a while, because of the huge over building and resultant surpluses of batteries from China, making LFP (and Lithium-ion for cars, too) dirt cheap.
I too have taken a detailed look at using storage to shift dispatchable (I.e. gas) a few hours.
One of the frequent claims is that shifting gas generation allows for increased efficiency. But any small increase in efficiency is more than offset by the storage losses due to round trip efficiencies.
Very thorough analysis, and thanks for putting a number on round trip battery energy round trip losses for charge and discharge. I hadn't seen that calculation before.
TANSTAFL - there ain't no such thing as a free lunch!
Another reason given for data centers using more storage, if I understand it right, is that a battery can 'start up' instantly, for immediate response for dispatch, whereas a gas turbine generator start up usually takes at least 5 minutes to get up to full load.
Batteries are capable of fast response IF they are operating on frequency control. The problem is, they need sufficient ENERGY (MWh) to sustain the required rates.
I am not aware of any storage at this time that is actually operating in full dispatchable mode.
Texas does pay for ancilliary services (https://www.ercot.com/gridmktinfo/dashboards/ancillaryservices), but these are not really necessary as more than sufficient inertia is available from thermal generation. It is really just a subsidy to pay storage facilities to keep them economically viable.
Gas turbines (simple cycle) that are shut down are not dispatchable. As you point out, they take time to start up and take up load. CCGTs take considerably longer.
When the grid operator knows a gas turbine generator will be needed for dispatchable service, it is started up and sync'd to the grid in droop mode. This allows it to run with zero load, but enables it to respond within seconds to maintain grid frequency.
Great article, great news. I would assume that the plans for additional natural gas plants would not include states with net-zero emission regulations, such as MD, NJ, VA.....but would include states not in RGGI WV, OH, PA. Can we determine where this additional capacity is planned?
That’s a great question. We’ll look into it! My initial search looks like there is some gas being built in Net Zero/RGGI states due to locational need. We may very well see more gas go to non-net zero or RGGI states over time.
I note that a consistent theme in the commentary here is that capital has been mis-spent in this two decades long renewables adventure and will require some time to correct. I could not agree more. The first step will be to convince unilities, public service commissions, and others to stop directing capital into more renewables.
Yes, natural gas is a high quality fuel and offers some design advantages when it comes to thermal power - the efficiency of the combined cycle plant in particular. The super-critical or near super-critical plants located near coal deposits offer some advantages also.
Harking back to the energy crisis of the 1960s and 1970s, quite a lot of that crisis resulted from a growing dependence on petroleum liquids for power generation; while at the same time depending on petroleum for transportation and chemical feedstock. I might also point out that in the latter 1960s utilities were looking at a compound growth of 6% in electric energy demand from around 1960 on which had them in a fury to plan new power plants. This brought with it worries about available water for cooling. That 6% growth did not continue, but the concerns over water did lead to innovation in plant designs.
We learned in that earlier energy crisis that a broad array of fuels available to us right here at home was a solution to the problem. I do have some concerns that we will repeat this history by becoming too reliant on natural gas for power generation; while depending on it also as a chemical feedstock and source of residential and commercial building heat with attendant issues over pipeline capacity and local storage.
Perhaps an analysis of the cost of replacing a 500 MW baselaod generator with solar plus storage might be enlightening, particularly if the plant must produce reliably through a storm such as winter storm Fern (~10 days).
The philosophical problem is that grid operators treat the intermittent generation issue poorly. If you were to contract with an electricity source that couldn’t tell you how much it would generate or when it would generate that would be a very stupid business transaction. Then when you did make that contract you had to make another transaction to backfill for the intermittent generation and pay for that backup to be ready without notice for an indefinite time period, that would only amplify your original stupidity. But then if you don’t care about cost or reliability you justify to yourself your lunacy. Such is the business philosophy of grid operators stacking more wind and solar into an already reliability stressed grid. Clearly we need to reboot and start over with a reasonable business philosophy and just perhaps that means different people working for the grid operators. Else you can just pay exhorbitant electric rates but only when the power is on.
"And thanks in large part to the explosive growth of data centers, reliability matters again, and that means wind and solar should learn to code."
Anyone else find it galling that when it was consumers, reliability didn't matter, but as soon as it affects the Tech Weenies, suddenly it matters again? I can explain it with an economic explanation of concentration of power/interest better able to represent/concentrate interest to governing bodies, but it's still galling and ought to be wrong.
Help a non-expert to understand. By 'storage', you (and the PJM graphic) mean LFP container-sized battery back up, right?
[I'm working on a piece about the increasing use of storage as a short term (couple hours) daily energy leveler. Even data centers are using storage to increase the efficiency of their gas turbine generator operation. It's feasible now, at least for a while, because of the huge over-production and resultant surpluses of batteries from China, making LFP (and Lithium-ion batteries for cars, too) dirt cheap.]
One other question - the phrase "interconnection queue submissions" confuses me. By "queue submissions" I have to assume you mean prioritizing who goes first, right?
Wouldn't it be better to just call it "interconnection queue prioritizing"?
as a resident of the J in PJM, I am delighted that they are finally understanding reality. of course, based on my electricity bills, I fear it is a little while before the benefits will be felt
It does not appear that the "J" governor understands reality or economics.
https://edreid.substack.com/p/consequences
Alas it has been that way for a while
Nice piece - fully agreed that the pendulum has swung with potentially a significant cost to Customers. The extreme focus on renewables over the past 10 years has given us a very expensive generation portfolio with less than desired reliability. The true cost has been both masked with all of the tax credits (transferred to federal government) and hidden by not fully recovering the costs of now retired assets (i.e., coal plants). Now as utilities and market pivot back towards a balanced generation portfolio and address the significant increase in demand driven by data centers/AI, utilities/IPP's are paying extremely high costs given the concentrated gas turbine demand.
- The cost of a spare turbine we've been evaluating increased by over 50% over the past 9 months.
- A recently received bid for a greenfield gas turbine resource was $1B (~20%) over the prior estimate.
Another way to think of it is that the mis-guided generation investment over the past 10 years driving elevated customer rates is going to take 10 to 20 years to resolve.
Another factor that will likely influence generation cost going forward is an issue that is frequently mentioned in the mining industry. The most attractive and lowest ore bodies have already been mined resulting in future mines requiring higher prices to justify economically. I see the same happening with the siting of new generation resources, especially here in the southwest with our limited water resources.
- Gas turbines being evaluated with expensive air cooled condensers as opposed to wet cooling
- New nuclear siting challenged by the need to identify sites with significant amounts of water to support meaningful plant capacities (2 x AP1000 requires ~44K acre-ft/year).
- Existing transmission is at or close to capacity resulting in new resources needing to add very expensive transmission as well.
- Attractive wind turbine sites that are closer to customer load have already been utilized.
In summary, cost of new generation will continue to increase independent of the soaring turbine costs.
Many good points.
“Say what you want about data centers.” Okay, I will. I have no problems with the technical analysis in this post. But what are the social and political costs of increased electricity use? I see the growth of data centers and AI causing massive increases in electricity demand. Corporate management and shareholders who already hold much of the wealth in our country benefit. Working class people, both in blue and white collar jobs, get laid off due to automation fueled by AI. The middle class foots more of the bill for electricity due to the expansion of data centers and AI. The share of the wealth in our country continues to shift from the lower and middle classes to the wealthy elites. Social unrest increases and political instability ensues. I hope I am wrong.
"The middle class foots more of the bill for electricity due to the expansion of data centers and AI. "
This does not have to be the case. There are several approaches available to isolate AI demand from the current grid, including "islanding" suggested by President Trump.
I see an intensive focus on ensuring the data centers/AI do not increase the cost for other customer classes. A number of ways this is being implemented:
- DC/AI customers fund the development of any incremental generation resources
- DC/AI customer absorb a meaningful portion of common costs (operational, transmission, and regulatory) incrementally reducing the amount allocated to other customer classes.
- DC/AI customers are required to agree to long-term contracts (20 to 25 years) with significant minimum payments (80% to 90% of planned load) as well as provide significant collateral in case they go out of business.
- DC/AI customers payments are in excess of the actual revenue requirement (additional security). If the customer does not sign a follow-on contract, this balance benefits customers and ensures they do not have to fund stranded generation assets.
Just my perspective from having visibility into multiple on-going DC/AI discussions.
I hope what you are saying is implemented. For now some communities are being hard hit by increasing electricity bills. And then there is also the issue of massive water usage by these data centers.
In the building phase thry will consume water same as any construction...but in the run phase the water is in a closed loop cooling system. The idea that DCs "consume" vast amounts of water is nonsense. It was dreamt up as a talking point by groups aligned to Neville Roy Singham. Please do a bit of reading on the subject before repeating tropes that may not be true. You can use AI to do that
Thanks for your input. I admit to believing DCs in general are huge water consumers without doing much research into the issue. I therefore did a deeper dive into the subject and am impressed by how much closed looped cooling can save water compared to evaporative cooling methods. According to one source I read most of the DCs currently use evaporative cooling. I hope closed looped is the default for future DCs. More than one article I read still found drawbacks to closed looped systems. There evidently is still water loss due to leaks and other issues but the closed looped beats evaporative cooling by a large margin. Another issue is that DCs are huge electricity consumers. If the electricity is supplied by plants using a lot of water (ie steam turbines), then the water usage issue is being relocated, not being completely mitigated. This is a fascinating subject.
Data centers need to generate and pay for their own power, and not burden the grid or consumers.
It's a little off topic, but you're right - the wealth gap just keeps getting bigger - and it's very hard on the working class, especially the working poor.
I have virtually no concern about income or wealth inequality
High income individuals get their income by providing goods and services that people want. They do not “take” money from the poor. They provide an option that people prefer. All unforced financial transactions benefit both the buyer and the seller, or the transaction would not happen. If a cup of coffee costs $4 and I buy it, it is because I value the coffee more than I value the $4. The seller in this case values the $4 more than the coffee.
I have many friends who are incensed at the wealth of Jeff Bezos. I am not. I buy most of the things I use from my local grocery store, Target, Kohls. Home Depot, Ace Hardware, Etc. They are close, they have what I want on hand, and I don’t generally have to make a special trip to shop at those stores. I buy things I purchase infrequently, things I can find at a store 45 minutes away (1.5-hour round trip just for the driving) or things I don’t know where to buy from Amazon. I get the item for what I would have purchased at a local store, if one exists, and I get something more valuable that I cannot buy for any price for free. That item is more time. I can’t buy a longer life, but I can reduce the time I spend acquiring some of the things I need or want. Thank you Jeff.
I say I have “virtually no concern” for a reason. A wealthy person can only do a few things with their money. They can:
- Buy goods and services themselves. The money spent there goes into someone else’s pocket, his or her local economy, etc. Wealthy people shopping benefits those they buy from and hundreds of other downstream people including many blue-collar workers. Those workers include people who mine materials, provide logistics, make subcomponents and finished goods and install or maintain any number of objects in a wealthy person’s home. All good.
- They can invest the money (probably the bulk of what they do). Investment grows the economy, increases productivity, and makes us wealthier. All good.
- They can pay taxes. People on the left would argue this is always good.
- They can donate money.
- Take money out of productive circulation. That includes money in their wallets, money in a safe at home, or I believe money in nonproductive assets like gold. I don’t know how large this pot is, but it is probably relatively small.
- Use their wealth to buy favors with government officials. I do not like this one. This one exists because our government is into too many things, makes too many rules, and interferes with our lives too much. A much smaller government would reduce the benefit of buying government favors.
The fact that someone else is rich, does not mean that I am poor or worse off. Imagine that the poorest among us all had $1 million a year in 2026 purchasing power but the richest had $1 trillion. Would we complain about the $1 million?
Income inequality is often inaccurately reflected by income numbers presented by the press. In the book “The Myth of American Inequality”, by Phil Gramm, he points out that incomes for the wealthy don’t reflect the taxes they pay (the top 20% of income earners pay almost all the income taxes) or the transfer payments that the bottom earners make. If I have $25,000 a year but get $10,000 in transfer payments, I am really making $35,000 a year. The wealthy have higher levels of family labor force participation and often work more hours.
Our understanding of things such as minimum wage are also wrong. The majority of minimum wage earners are younger than 25 or retired. They are not a family of 4 trying to survive on a minimum wage job. Young earners need entry jobs to gain skills which help them over their economic life’s. Minimum wages laws make those skill hard to acquire.
If we want to help the poor, we need to do things such as:
- Eliminate the minimum wage - Get rid of licensing for most occupations - Allow school choice so we can improve education benefiting all Americans - Kill the idea of college for all. Something like 40% of people who enter college never graduate. That is time and money poorly spent. I am not opposed to post high-school education but a degree in gender studies is financially worthless. Getting skills as a welder may translate to much better long term financial success.
I am concerned about wealth inequality in the United States since the trend has been a reduction of wealth for the middle class and an increase for the upper classes. The standard of living has worsened for the lower and middle classes. This is bad for the country for a number of reasons. Poorer health outcomes for the uninsured, housing affordability crises, capture of the government by wealthy elites. I could go on. I’m not anti-capitalist. But I think we can have a robust safety net for people without taxing the wealthy so heavily that they are disincentivized to work hard and innovate. I’m not a fan of trickle down economics. If the lower and middle classes are given a decent wage then doesn’t that benefit the economy as much as the uber wealthy benefiting “people downstream,as you put it, when they buy goods and services? Henry Ford evidently paid his workers a good wage so they could afford to buy cars which ultimately benefited Ford Motor Company. That’s trickle up economics.
The shift to natural gas combined-cycle generation will quickly require the addition of natural gas pipeline capacity to provide firm gas deliveries to those powerplants and perhaps also the addition of natural gas underground storage capacity.
Very accurate statement - development of the Desert Southwest Pipeline is a key enabler to add much needed gas generation in the southwest, at a current cost estimate of $5.6B.
Yep and in New England specifically, Massachusetts the 1600 megawatts of “offshore” wind have laid a giant egg and it isn’t golden. All these states and ISO’s responding to the mid wit politicians who pushed the solar and wind agenda are now seeing (whether they will admit failure or not is another story) cause and effect of the failure of the policies they pushed. Rate payers in New England will need to hold onto their hats as the python digests the mess of a renewable rabbit these fools created. Of course at the moment they are saying mess, what mess? Move along. Folks nothing to see here. It will get worse for the rate payers before it gets better, but like anyone kicking a bad habit it starts with recognizing the habit is bad and needs to stop. We are getting close to that point here in NEISO. Great post EBB’s!
For years we were told that for human needs reliability was not a big deal. Now that the big media company machines need reliability, it sits center stage. It is interesting how it helps to set priorities.
The demand for storage is interesting. But it agrees with my earlier statements that batteries are no longer a renewable item, and now simply a power system accessory. In this capacity they work well with fossil fuel generators, absorbing power when loads are low to keep units at optimum heat rate, discharging when loads are high to supplement generation.
I wish people would stop referring to Storage POWER Ratings (MW). Power Ratings are the capability to discharge electricity IF THEY HAVE THE ENERGY AVAILABLE.
Although important POWER is secondary to ENERGY CAPACITY (MWh). This is what tells you the rate & duration of discharge. In other words, the ability to meet demand after Wind and Solar fail.
All the POWER (MW) Capacity is meaningless if you do not have sufficient ENERGY to supply the POWER.
To help understand this, look at California Energy Stoage Performance: (Fig 3a & 3b) https://wrjohn1.substack.com/p/california-caiso-mar-2026-the-path.
California is the poster child for energy storage, but it's capability to displace hydrocarbons is insignificant.
Translation: “battery storage systems are not dispatchable or baseload systems.”
Exactly!
You're right, of course, as far as storage being capable of helping in any long duration wind and/or solar drought.
However, I'm working on a piece about the increasing use of storage as a short term (couple hours) daily energy leveler. It's feasible now, at least for a while, because of the huge over building and resultant surpluses of batteries from China, making LFP (and Lithium-ion for cars, too) dirt cheap.
I too have taken a detailed look at using storage to shift dispatchable (I.e. gas) a few hours.
One of the frequent claims is that shifting gas generation allows for increased efficiency. But any small increase in efficiency is more than offset by the storage losses due to round trip efficiencies.
You can see my analysis here: https://wrjohn1.substack.com/p/can-energy-storage-reduce-nat-gas
Thanks for the link, Bill!
Very thorough analysis, and thanks for putting a number on round trip battery energy round trip losses for charge and discharge. I hadn't seen that calculation before.
TANSTAFL - there ain't no such thing as a free lunch!
Another reason given for data centers using more storage, if I understand it right, is that a battery can 'start up' instantly, for immediate response for dispatch, whereas a gas turbine generator start up usually takes at least 5 minutes to get up to full load.
Batteries are capable of fast response IF they are operating on frequency control. The problem is, they need sufficient ENERGY (MWh) to sustain the required rates.
I am not aware of any storage at this time that is actually operating in full dispatchable mode.
Texas does pay for ancilliary services (https://www.ercot.com/gridmktinfo/dashboards/ancillaryservices), but these are not really necessary as more than sufficient inertia is available from thermal generation. It is really just a subsidy to pay storage facilities to keep them economically viable.
Gas turbines (simple cycle) that are shut down are not dispatchable. As you point out, they take time to start up and take up load. CCGTs take considerably longer.
When the grid operator knows a gas turbine generator will be needed for dispatchable service, it is started up and sync'd to the grid in droop mode. This allows it to run with zero load, but enables it to respond within seconds to maintain grid frequency.
Great article, great news. I would assume that the plans for additional natural gas plants would not include states with net-zero emission regulations, such as MD, NJ, VA.....but would include states not in RGGI WV, OH, PA. Can we determine where this additional capacity is planned?
That’s a great question. We’ll look into it! My initial search looks like there is some gas being built in Net Zero/RGGI states due to locational need. We may very well see more gas go to non-net zero or RGGI states over time.
I note that a consistent theme in the commentary here is that capital has been mis-spent in this two decades long renewables adventure and will require some time to correct. I could not agree more. The first step will be to convince unilities, public service commissions, and others to stop directing capital into more renewables.
Yes, natural gas is a high quality fuel and offers some design advantages when it comes to thermal power - the efficiency of the combined cycle plant in particular. The super-critical or near super-critical plants located near coal deposits offer some advantages also.
Harking back to the energy crisis of the 1960s and 1970s, quite a lot of that crisis resulted from a growing dependence on petroleum liquids for power generation; while at the same time depending on petroleum for transportation and chemical feedstock. I might also point out that in the latter 1960s utilities were looking at a compound growth of 6% in electric energy demand from around 1960 on which had them in a fury to plan new power plants. This brought with it worries about available water for cooling. That 6% growth did not continue, but the concerns over water did lead to innovation in plant designs.
We learned in that earlier energy crisis that a broad array of fuels available to us right here at home was a solution to the problem. I do have some concerns that we will repeat this history by becoming too reliant on natural gas for power generation; while depending on it also as a chemical feedstock and source of residential and commercial building heat with attendant issues over pipeline capacity and local storage.
As always, time will tell.
Perhaps an analysis of the cost of replacing a 500 MW baselaod generator with solar plus storage might be enlightening, particularly if the plant must produce reliably through a storm such as winter storm Fern (~10 days).
In Texas, wind power has destabilized our once very reliable grid.
The philosophical problem is that grid operators treat the intermittent generation issue poorly. If you were to contract with an electricity source that couldn’t tell you how much it would generate or when it would generate that would be a very stupid business transaction. Then when you did make that contract you had to make another transaction to backfill for the intermittent generation and pay for that backup to be ready without notice for an indefinite time period, that would only amplify your original stupidity. But then if you don’t care about cost or reliability you justify to yourself your lunacy. Such is the business philosophy of grid operators stacking more wind and solar into an already reliability stressed grid. Clearly we need to reboot and start over with a reasonable business philosophy and just perhaps that means different people working for the grid operators. Else you can just pay exhorbitant electric rates but only when the power is on.
How many grid operators are forced by "must take" laws at the state level?
"And thanks in large part to the explosive growth of data centers, reliability matters again, and that means wind and solar should learn to code."
Anyone else find it galling that when it was consumers, reliability didn't matter, but as soon as it affects the Tech Weenies, suddenly it matters again? I can explain it with an economic explanation of concentration of power/interest better able to represent/concentrate interest to governing bodies, but it's still galling and ought to be wrong.
Great article - thanks.
Help a non-expert to understand. By 'storage', you (and the PJM graphic) mean LFP container-sized battery back up, right?
[I'm working on a piece about the increasing use of storage as a short term (couple hours) daily energy leveler. Even data centers are using storage to increase the efficiency of their gas turbine generator operation. It's feasible now, at least for a while, because of the huge over-production and resultant surpluses of batteries from China, making LFP (and Lithium-ion batteries for cars, too) dirt cheap.]
One other question - the phrase "interconnection queue submissions" confuses me. By "queue submissions" I have to assume you mean prioritizing who goes first, right?
Wouldn't it be better to just call it "interconnection queue prioritizing"?
Several PJM states are also members of RGGI and might well oppose the expansion of natural gas generation in PJM.
ReNewable Energy Questions
1 Why do renewables need subsidies ?
2 Why are electricity rates increasing ?
3 Why are there more black/brownouts with renewables?
“We get a tax credit if we build a lot of wind farms.
THAT'S THE ONLY REASON TO BUILD THEM.
They don't make sense without the tax credit.”
- Warren Buffett