Many environmentalists are claiming that climate change is an existential threat. Organizations like the Extinction Rebellion scream that global warming will end civilization. It’s fine to raise awareness, but better to analyze problems and propose workable solutions.
“Renewables” are a bandwagon slogan for wind and solar energy sources to replace CO2-emissions from burning fossil fuels. Global investment in renewable energy reached $282 billion in 2019 alone, with hopes that future energy storage technology can solve the intermittency issue.
Can we electrify everything with renewable energy from water, wind, and solar sources? Is the green new deal feasible? What would be the cost? What about wind lulls, cloud shading, and nightfall? Can we store enough intermittent energy?
isThis is a simple model of a competitive market. The buyer would be the local utility company that distributes electricity to homes and businesses. The seller might be a natural gas power plant, a hydro plant, a wind farm, or PV solar array. Money and energy are exchanged at a typical price of $0.05 per kWh, with the buyer choosing power from the least expensive supplier. Unfortunately the “market” doesn’t work this simply.
Feed-in tariffs. Legislators and regulators often dictate where the energy should be purchased. A decade ago feed-in tariffs were established. Tariff prices were set by regulators at levels high enough to encourage wind and solar developers to participate. Whenever the wind blew or the sun shone, the utility was required to buy the wind/solar generators’ electricity at the dictated price, above the competitive market price. Less expensive power being generated by full-time producers was curtailed.
Renewable portfolio standards. Some states require utility companies to obtain enough wind-sourced or solar-sourced energy to compose a dictated minimum percent of all electric power purchased and distributed. That is the renewable portfolio standard. Usually the required percentage increases in future years, say 50% by 2050, leading to sloganism such as “50 by 50”. Many utilities and their suppliers do not have power line access to such wind/solar sources, so regulations let the utilities buy the emission-free property of wind/solar power generated far away. Such a distant generator sells the kilowatt-hours of energy locally, but is also awarded a renewable energy credit (REC) that represents the emission-free property of the power generated. That REC can be sold to the utility that needs the wind/solar energy to meet its states’ renewable portfolio standards. There are multiple classes of RECs: for PV solar, for thermal solar power, for big hydro, for small hydro, for off-shore wind, etc. RECs are sold at regular auctions in New Jersey, Massachusetts, and other states. The 2020 New Jersey solar REC price is $220/MWh, or $0.22/kWh. Note that such REC revenue is 4X the typical revenue from selling the energy.
When you see headlines claiming record-low prices for solar power, such as $0.02/kWh, understand that the developer may also receive an added $0.22/kWh for RECs. That money is paid out by a utility company; it’s charged to the rate base, the pool of costs that are spread out to all customers, raising prices for utility customers a bit. Paying for such RECs adds to the price of electricity in many areas of the US, masked by the fact that costs of electricity generated by fracked natural gas have dropped by almost half over the last years.
Many members of the public and the media anticipate that wind and solar power will deliver ample clean energy and solve the climate crisis. Legislators and regulators have obliged and created incentives. In the US the laws and rules are instituted by many overlapping governing bodies, such as Congress, the Federal Energy Regulatory Commission, state legislatures, cities, towns, and regional organizations that control dispatch of electricity from selected generating suppliers. The various subsidies differ by region and by year. The rules are complicated, changing, uncoordinated, poorly understood, and exploited by clever energy businesses, as was the intent of the incentives. Below are descriptions of the money flows available to developers.
Electricity generated. Consumers pay utilities who pay generating suppliers for kilowatt-hours delivered, $/kWh. In some regions this sale of energy is less than half the revenue received.
Electricity curtailed. Feed-in tariffs specify that wind/solar generators have first priority to deliver their energy, but sometimes the operating thermal plants (gas, coal, nuclear) can’t reduce generated power rapidly, so the dispatching organization curtails the generated wind/solar energy, but they must be paid as if they did generate the electricity. The dispatching organization charges the slow-reacting thermal plants for this cost of paying higher prices for curtailed wind/solar generation, creating negative prices for thermal plants’ electricity.
Renewable energy credits were discussed under renewable portfolio standards.
Production tax credits. Wind/solar generators accumulate credits of $0.023/kWh for each kWh of electricity delivered. US income tax credits can be transferred to profitable organizations, so the credits are as good as cash. There are different production tax credits for geothermal, nuclear, landfill gas generators according to a schedule.
Green bond discounts. A green bond is a fixed-income instrument designed specifically to support specific climate-related or environmental projects. Green bonds often come with tax incentives or guarantees to enhance their attractiveness to investors.
Grants. Charitable organizations or government bodies sometimes grant money to defray costs of wind/solar projects.
Property tax exemptions. States, cities, and towns sometimes reduce property taxes for wind/solar projects.
State income tax credits. Vermont, for example, grants solar project developers a tax credit of 30% of the capital cost of building solar panel generators.
Accelerated depreciation. In vanilla tax accounting, you deduct froim taxable income the expense of a $20 million capital project with a 20-year expected useful life, at $1 million per year, for 20 years. With accelerated depreciation, much of the costs can be deducted in the first few years, reducing taxes enough to reduce borrowing costs.
Federal income tax credit. A federal income tax credit of 30% helps to subsidize solar panel energy generation. Large generating suppliers can not take both this credit and the production tax credit.
Capacity reserve. For their consumers, utilities rationally want assured future supplies of electricity from their generating suppliers. The regional dispatching organization negotiates supplier commitments to deliver electricity in future months and years, and will pay suppliers to make such promises, so suppliers build the plants and contract for fuel. For example, a natural gas electricity generating company buys enough future capacity in a pipeline from Texas to transport the gas to its power plant in New England. The rules are very complex and disputed, partly because wind/solar generators are awarded capacity reserve payments even though they can not promise the wind will blow and the sun will shine.
Meredith Angwin’s new book, Shorting the Grid: the Hidden Fragility of our Electric Grid, raises the specter of blackouts from irresponsible regulation.
The US National Renewable Energy Laboratories is a US DOE organization chartered to promote “renewable” wind, solar, hydro, biomass, landfill gas, geothermal energy sources. Their very existence and continued funding depends on there being a favorable role for these sources, so I expect their cost estimates are not overstated.
CAPEX. Column 2 deals with the initial, capital expenses (CAPEX) for wind turbines on land with a nominal, name-plate power generation capability of 2.4 megawatts at optimum wind speed. NREL estimates CAPEX of $1,470/kW of generating capacity.
Column 3 presents the allocated cost of paying back that CAPEX at 7.5% over 30 years, $30.3/MWh.
OPEX. Column 3 also presents the ongoing operating costs (OPEX) of $12.1/MWh.
The resulting NREL estimate is $42/MWh for land-based wind turbines. The assumed lifetime of 30 years and the capacity factor of 42.5% may well be optimistic.
LCOE stand for levelized cost of energy, cost of the electricity delivered, ignoring when it was delivered. To be commercially useful, electricity must be available when needed. Wind power is subject to wind lulls, which can last for a week. Wind power sources must be matched with dispatchable sources that can be turned on/off as the wind lulls/blows. Typically the source is a natural gas turbine-generator, whose capital cost of approximately $1,000/kW is never included in the wind cost estimates.
Here’s a similar analysis for off-shore, seabed-mounted larger, 5.5 MW nameplate capacity wind turbines. The assumed capacity factor is higher, but not enough to make up for the higher construction cost, resulting in the $89/MWh energy cost. I believe the 30 year life exposed to the ocean and storms to be optimistic.
This plot is the result of measuring actual capacity factors of actual wind turbines. The prior NREL estimates of 42% and 49% seem to be optimistic.
Lazard is a sell-side financial analysis firm whose energy cost estimates are often quoted in media. Sell-side analysts help companies sell investments. Buy-side analysts help find stocks and bonds for investors’ portfolios.
Highlighted in red are summaries of Lazard’s estimates of costs for electricity from utility-scale solar and wind generators. Such prices ~ 4 cents/kWh are a bit lower than typical prices of energy sold to utilities. Lazard estimates are low in comparison to what utilities pay.
For example, in Vermont the utility companies are directed to buy electricity at ~ 11 cents/kWh from new solar panel farms. Older ones are paid ~ 20 cents/kWh. The utility is forced to buy electricity at ~ 21 cents/kWh from net-metered, customer-owned, roof-top solar arrays. The costs of solar payouts are charged to the rate base and then shared among all customers.
The largest contributor to the cost of electricity is the capital cost of the power plant. Knowing the $/MW of capacity, interest rate, and number kilowatt hours sold to repay the capital cost, the PMT spreadsheet function computes the capital cost per kWh.
- Deepwater Wind is a wind farm being built off the coast of Rhode Island, US. Ivanpah was a concentrated solar thermal power plant in California which did not come close to expectations.
- Topas, Agua Caliente, and Setouchi are all photovoltaic solar panel powered plants.
- Noor is a concentrated solar power plant under construction in Morocco; it includes storage of heat energy in tanks of molten salt, so that thermal energy can be saved then used to generate electricity after the sun sets.
- The Westinghouse AP1000 is a nuclear power plant project in Georgia; the project has severely overrun its planned costs.
- ThorCon is a proposed 4th generation fission power plant being designed for Indonesia.
- Riau is a coal-fired plant being constructed in Indonesia.
The capital costs were obtained from published media reports. Operating costs are rough estimates. The Total cost per kWh are the minimum electricity prices that must be charged for an unsubsidized power plant project to be economically feasible.
Government guarantees may permit the developer to borrow money at costs lower than 8%; guarantees are a form of subsidy, exercised in the Ivanpah example.
All these examples of computed economic costs for wind and solar sourced power exceed optimistic estimates published in media reports. Published project announcements rarely include the project’s cost per kWh, but you can compute the minimum cost if the capital cost can be determined.
The initial deal with National Grid funds a power plant costing $6.67 per watt of generating capacity. For comparison, typical natural gas turbine plants cost $1/watt. The 24.4 cents/kWh may seem high, but not compared to the cost of island power plants running on diesel fuel oil barged in.
The full project is claimed be more economical at $3.90/watt.
The true costs of yet another off-shore wind power project are being kept secret. The bids are said to be published, but in fact they are redacted to meaninglessness. The consuming public will ultimately pay the costs via electricity rates, but are denied any visibility to the dealings between utility commissioners, regional dispatching organizations, utility companies, and project developers.
If the costs were low, wouldn’t they be proudly published?
The URL above has since been changed and now links to this message, “SITE DELETED This site has been deleted by the owner.”.
Because wind and solar power generation sources are unreliable, additional generating plants must be built to start up quickly and provide power when there is no sunshine or wind. Countries like Denmark have successfully used much wind power by supplementing it with hydropower from Sweden, but most hydro potential has already been put into service. Renewables backup service is mostly from natural gas power plants, now the largest source of electricity in the US.
The natural gas industry promotes renewables because they create opportunities to build and operate new natural gas power plants. The graphic above contains their advertisement.
There are two kinds of natural gas turbine generator plants. The least expensive is a generator turned by a turbine engine similar to one in a jet plane. It has the advantage of starting up quickly, so it can replace fading wind or solar supplied energy. It’s a natural gas combustion turbine (NGCT).
The most efficient combined cycle gas turbine (CCGT) achieves an admirable thermal to electric energy conversion efficiency of up to 63%. The hot exhaust gasses from a natural gas combustion turbine pass through a very large heat exchanger to make steam that powers another turbine-generator. A CCGT takes longer to start up and is more expensive. During power ramp up or ramp down it is less efficient.
The natural gas power plants that back up wind and solar are NGCTs or CCGTs operating in less efficient modes.
From a system point of view, wind turbines increase CO2 emission. Suppose a utility needed to generate 1000 MW of electricity. It has a choice between (a) wind turbine farm with NGCT power plant backup or (b) efficient CCGT plant running at full capacity. Here’s an illustrative example.
Case (a) the NGCT runs 70% of the time with an electric/thermal efficiency of 39%, using 1000/0.70/0.29 = 2410 MW of thermal energy from burning natural gas. The wind turbine runs 30% of the time burning no fuel.
Case (b) the CCGT runs 100% of the time with an electric/thermal efficiency of 60%, using 1000/0.60 = 1670 MW of thermal energy from burning natural gas. The wind-turbine-NGCT choice emits 44% more CO2. CO2 emissions are proportionate to gas burned, so 2410/1670 = 144% of the CO2 of case (b).
In a real example, such as the island of Ireland in 2014-2015, electric power was provided by CCGT generators plus wind turbines, without interconnections to UK or Europe . Ireland was powered by CCGTs that were turned down when the wind blew, reducing natural gas burning and CO2 emissions. Ireland recorded fuel consumption and power output for its grid every 15 minutes. Powering the CCGTs up/down dropped their thermal-electric efficiencies from ~ 50% to 32%, so they emitted more CO2 per MW of power generation. Adding 2 GW of wind power to the Irish grid reduced CO2 emissions equivalent to shutting off 1 GW of steady-running CCGTs. References: Wheatly, Mearns, Udo.
In the US the EIA reports the full fleet of both CCGTs and NGCTs averaged a measured efficiency of 37% for 2016 and modeled efficiency of 37% in 2020.
Natural gas is the largest source of electric power in the US. Here’s an example from ISO-NE (Independent System Operator, New England).
Note that refuse burning can consume much of our waste stream while generating electricity. Burning wood destroys forests and depletes soils. Do explore iso-ne.com.
Fracking has reduced the costs of extracting oil, propane, and natural gas. Consequently electricity from burning natural gas has dropped by half.
Question: why hasn’t your electric bill dropped proportionately?
About 2/3 of electric power demand is constant. The baseload can be economically provided by power plants designed to run full-time at constant temperatures. These include coal-fired plants, CCGT plants, and nuclear power plants. Powering them down during sunny or windy periods can save fuel, but reduces efficiency during the transistion; thermal shock shortens plant lifetimes.
Modern wind turbines are enormous construction marvels. This 12 MWe GE Haliade X is designed for installation on the seabed. To the right is a scaled comparison to a 500 MWe liquid fission power plant designed by ThorCon.
The lefthand vertical scale denotes CO2 emissions in grams per kilowatt hour of energy delivered. Look for the little CO2-cloud icons on the plot. The CO2 emissions include those emitted during the construction of the power plant.
Various generating technologies are represented by the six colored bars. Colors denote the natural resources used to build a power plant capable of generating one terrawatt hour of electric energy per year (a 114 MW power plant).
Wind turbine power plants use vastly more concrete and cement than nuclear plants. Solar PV power plants use much more steel and iron.
A friend from Stanford thought that energy measures such as quads and terawatt-hours were not meaningful to the general public, so he wrote a book and introduced the idea of the expressing large quantities of energy in units of burning a cubic mile of oil (CMO). Here’s Ripu’s talk.
Burning 1 CMO per year yields ~ 5000 GW of power. World total energy consumption is a bit under 4 CMOs per year.
Equaling the power from burning 1 CMO per year would require 4.2 billion roofs with 2.1 kW PV solar panels generating on average 20% of the max full-sun capability.
Installation of that many roofs would take 50 years at a rate of 250,000 roofs per day.
A CMO of energy per year could be generated by 7,700 solar concentrated solar parks of 900 MW operating at 25% availability.
We’d have to build 3 CSP parks more than twice the size of the Ivanpah failure every week for 50 years.
Concentrated solar power plants were promoted in the 2000s, but expensive failures such as the 392 MW Ivanpah CSP plant have discouraged investments. Photovoltaic solar panels are now less expensive than CSP plants, but CSP plants can potentially be augmented to store the collected heat in tanks of molten salt to enable generation after the sun sets.
It would take 200 large dams delivering 50% of 18 GW to deliver the hydropower equal to burning 1 CMO per year.
The Three Gorges Dam in China has a max capability to generate 22 GW; on average it generates half of that. Building that dam took 14 years and the flooding displaced over a million people.
We’d have to build 1 new 18 GW dam every 3 months for 50 years to match the power of burning 1 CMO per year.
Three million wind turbines could match the power of burning 1 CMO/year, but it would take 50 years.
This is an example of what 100% renewable energy would require be built. However, it would not work because wind and solar are intermittent energy sources, not providing power when needed.
Suppose a wind sourced electricity generator has a 33.3% capacity factor. Tripling the number of wind turbines will not create a 100% reliable power source. Weather systems can be half a continent wide, lack of wind at one place may well mean there is no wind anywhere within transmission line distance.
Here is an example of a 100-hour lull. During this time Germany had to depend on reliable, dispatchable power from coal, natural gas, diesel, or hydro sources. With 100% renewable energy there would have been no power for four days.
Intermittency is an unsolvable problem with wind and solar energy sources.
Battery prices are indeed dropping, but nowhere near enough to be affordable solutions to lulls in wind or solar power. A Tesla Powerwall costs $7000 before installation and stores 13.5 kWh of energy. That’s $518/kWh of storage, plus installation costs. Media reports claim prices will soon drop to $100/kWh. That’s unlikely. That’s not enough to be affordable.
Storing just 1 day of world 3000 GW power would require 36 billion Powerwalls. This could take 10 years building 1000 units per second.
The Energy Information Agency of the US Department of Energy studied battery storage systems and reported $772 per kWh of storage capacity.
This MIT study looked just at New England power over a full year. Storing enough power to get through the winter would be way too expensive. This cost might be partially reduced by installing enough excess generation capacity to get through winter’s shorter days.
Ripu’s book and talk continues on to compare burning a cubic mile of oil to a fleet of nuclear power plants. To many people, building one 900 MW electric power plant every week is outrageous. But this is the objective adopted independently a half-decade ago by a small group of environmentally conscientious engineers who founded the ThorCon companies.
International commerce uses fleets of container ships and oil tankers. This has created a shipbuilding industry, with multiple competitors able to build large steel vessels quickly and competently. The industry is based in Asia.
- Hyundai Heavy Industry, Ulsan, South Korea, 1428 ships
- Daewoo Shipbuilding – Okpo, South Korea, 834 ships
- Samsung Heavy Industry – Geoje, South Korea, 785 ships.
- Hyundai Samho – Samho, South Korea, 372 ships.
- Mitsubishi Heavy Industry – Nagasaki, Japan, 315 ships.
A single, large shipyard has the capacity to build 40, 500-MW ThorCon power plant vessels per year. The ThorCon objective is to utilize shipyards to build a production system for power plants.
In addition to the need to satisfy 3000 GW of current world electric power demand, we need to supplant 574 GW of the developing nations’ planned coal-fired power plants.
Convincing the developing nations to choose liquid fission source electric power is feasible, IF the zero-emission, reliable power plants are cheaper to build and operate than coal-fired plants. Once developing nations are powered up with cheap, high-quality power they will become internationally competitive, pressuring OECD nations to cut energy costs using liquid fission power.
The intermittent nature of wind and solar power means it can not provide electricity on demand. Reliable power is essential for industry, commerce, infrastructure, etc.
A wind and solar power system requires a redundant power generation system that can supply 100% of the power demand. The cost of that redundant power system is not included by advocates presenting the $/kWh cost of wind and solar.
Wind and solar deliver only ~ 1/3 of the energy demand, because of intermittency.
The total cost of batteries for storing just one day of world energy needs is $250 trillion.
Because the energy density of harvested sunshine and wind is low, the wind turbines and solar panels are large and spread out, requiring many more materials and natural resources than do fission power plants.
In many US jurisdictions, the revenue from selling kilowatt-hours of energy are smaller than revenue from subsidies such as the production tax credit and renewable energy credits.
In the instants when when the wind is blowing or the sun is shining the real costs (LCOE) for the generated energy are somewhere between 4 cents/kWh (Lazard) and 9 cents/kWh (NREL).
Return to Electrifying Our World.