Belief that bigger is better — or, at least, a lot cheaper — helped sideline nuclear power. It now imperils wind power.
In July, at the first commercial-size U.S. wind farm under construction, near Martha’s Vineyard, a turbine blade split apart and fell into the Atlantic. In May and again in August, at the new Dogger Bank wind farm 80 miles east of England, blades broke off their towers and slammed into in the North Sea.
All three failures involved a mammoth new wind turbine design from GE Vernova, optimized to abundant offshore winds. The 13-megawatt turbines, dubbed Haliade-X by GE Vernova, a spinoff from the old General Electric conglomerate, are the world’s largest, and are nearly four times more powerful than the average wind turbine installed in the U.S. last year, and a full order of magnitude (10x) beyond typical windpower machines installed 20 years ago, according to U.S. Department of Energy data.
To be sure, dozens of other turbines at both wind farms are operating trouble-free and helping displace fossil-fuel generation, although the Vineyard Wind farm is now shut. The expected output of each 13-MW unit, 56 million kilowatt-hours a year, will let power grids pull back on incumbent fossil-fuel power plants that would otherwise spew 25,000 tons a year of climate-disrupting CO2, while maintaining wind energy’s meteoric growth. In Great Britain, wind turbines now stand neck-and-neck with power plants burning “natural” (methane) gas as the top electricity source, providing a third of Britain’s electric generation in the first quarter of 2024, according to Energy Advice Hub. In the U.S., although wind power’s kilowatt-hour production fell by 2 percent in 2023 — the first year-on-year slip this century — wind turbines are producing 40 times more electricity than two decades earlier and now account for 10 percent of the nation’s electricity generation.
Despite, or perhaps because of, wind power’s growing prominence, the concatenation of the three incidents is worrisome. The Vineyard breakage seemed to validate the fears of East Coast commercial fishers and other objectors to the large-scale offshore windpower development that has become a linchpin of regional and national drawdowns from fossil fuels. “Jagged pieces of fiberglass and other materials from the shattered blade drifted with the tide, forcing officials to close beaches on Nantucket,” the New York Times reported this week.
Indeed, the Times headlined its story on the Vineyard mishap, “Broken Blades, Angry Fishermen and Rising Costs Slow Offshore Wind,” adding the subhead, “Accidents involving blades made by GE Vernova have delayed projects off the coasts of Massachusetts and England and could imperil climate goals.”
What happened?
Why the blades broke apart isn’t yet clear. According to the Times’ story, GE Vernova has labeled the three incidents “one-offs rather than systemic flaws … but has provided few details about their causes.” The first failure at Dogger Bank, in May, resulted from “an error during installation,” a company spokesperson told the paper, while the second, in August, “happened because a turbine was left in a ‘fixed position’ during a storm.” Trade publication Maritime Executive reported that GE Vernova told investors that the Vineyard blade rupture resulted from a “manufacturing deviation” in the bonding of the blade at a Canadian production facility. But the company “declined to confirm any details about the blade[s] that failed at the UK wind farm and if [both] came from the same manufacturing facility,” the publication said.
[On Sept. 20, just hours after we posted this story, GE Vernova said it planned to downsize its offshore wind business with 900 job cuts, many at the company’s turbine factory in Saint Nazaire, France.]
The cause could be rooted in the sheer size of the blades and, perhaps, the rapidity with which the industry has scaled up. The blades on the Haliade-X offshore wind turbine are 50 percent longer than those on a representative 3.3-MW land-based wind turbine: 220 meters tip-to-tip for the 13-MW model, according to GE Vernova, vs. 148 meters for the 3.3-MW machine, per the U.S. National Renewable Energy Laboratory’s 2022 Cost of Wind Energy Review (pdf). The disparity in bulk is probably well over 2 to 1; if the blade shapes are the same, the ratio of their areas would be 1.50 squared, i.e., 2.25 to 1. If the bigger blades are thicker as well, their weight could be triple that of the conventional blades.
It may be that current technology can’t mass-produce such enormous objects to the quality required to withstand the stresses from constant rotation. Even slight manufacturing defects that smaller turbines could handle might be unforgiving for giantic blades.
This isn’t to say that advances in metallurgy, material bonding and non-destructive testing couldn’t restore reliability for giant wind turbines in the future. The checkered history of rapid size-scaling in the U.S. nuclear power industry may be instructive.
Boosting reactor sizes proved disastrous in the U.S.
Throughout U.S. nuclear power’s “bandwagon era” — circa 1957-1974 — the U.S. Atomic Energy Commission and reactor manufacturers subscribed to the idea that nuclear costs would enjoy pronounced economies of scale. Their rule of thumb was that any doubling of reactor capacities — from 100 to 200 megawatts (for so-called “pilot” plants) and, later, from 500 to 1,000 MW — should raise costs only around 50 percent, tantamount to a 25 reduction in per-kW costs.
(The 50 percent cost rise would equate to a factor multiple of 1.50. Dividing that by two, for the doubling in megawatts, would yield a per-kW cost multiple of 0.75, i.e., a 25 percent drop.)
This impressive scale-economy made sense — on paper. Costs tend to track equipment surface areas, while capacity is proportional to reactor volume, portending only a 60 percent rise in costs per doubling of capacity. (Mathematically, two raised to the two-thirds power is roughly 1.6. Why two-thirds? Because surface area rises with the square of length while volume rises with the cube.) This would dictate a 20 percent reduction in per-kW costs from doubling plant capacity. Other cost elements like siting, permitting, engineering, and project mangement would, it was thought, display steeper economies, lifting the overall per-kW cost reduction per doubling of capacity to around 25 percent.
These upbeat expectations from reactor upsizing motivated successive doublings in reactor capacities through the 1960s and into the 1970s. The size increases did cut per-kW costs (or, at least, they helped hold back the tide of increased costs to comply with increasingly stringent safety regulations), but with diminishing returns. My own empirical analysis of costs to complete U.S. reactors, published in 1981, found only a 13 percent drop in per-kW costs per doubled reactor size, even when controlling for so-called regulatory creep. That saving was only half as great as the AEC had posited. Worse, larger reactors took far longer to build than smaller ones, which tied up vast amounts of capital, postponed nuclear displacement of fossil-fuel electricity, and spooked investors.
Larger reactors also proved harder to keep in service, their “teething” problems sometimes persisting for decades. That troublesome era is now decidedly in the past. The U.S. nuclear power sector’s average “capacity factor” has climbed steadily from the post-Three Mile Island accident trough of 55-60 percent operability to nearly 90 percent since around 2000. Still, with investor losses, high utility bills and excess carbon emissions, the toll from too quickly upsizing U.S. nuclear power plants proved immense.
A mid-range carbon price would match cost savings from doubling wind turbine sizes
The scale-economy curve for wind power in the graph above isn’t statistically stout, having been extrapolated from a mere two data points for land-based turbines in the NREL Wind Energy Cost report referenced earlier. (Readers with additional data: please share it with us!) That said, it conveys a message: double the size of an individual wind turbine and the per-kW capital cost should diminish by 18 percent. Factor in greater productivity — I assume that each kW of capacity of the larger turbine produces 9 percent more kWh’s than the smaller one — and the overall cost per kWh of wind power (“levelized cost of electricity,” in industry parlance) falls by 25 percent with a doubling of the turbine’s megawatt size.
That’s no small saving for supersizing, though of course it requires that projects employing the extra-large turbines actually come to fruition. According to the most recent (Aug. 19) Nantucket Town and County 2024 Turbine Blade Crisis Updates page — the name alone is a telling indicator — all installation and operation of turbine blades for the Vineyard Wind project are on hold, though placement of towers and nacelles (the equipment-bearing structures topping each tower) is permitted.
For Vineyard Wind, then, and perhaps for the Dogger Bank project as well, paper savings from going large have become a cruel joke, at least for the time being. Putting that aside, I’ve calculated the theoretical carbon price that would raise the sale price of wind electricity by the same amount that a halving of turbine capacity raised its all-in cost. Rephrased as a question: How big of a carbon price would have to be baked into the cost of prevailing fossil-fuel electricity — assumed to be from the mainstay of the U.S. power system, a combined-cycle power plant burning methane gas — to compensate for sticking with prior 6-7 MW sized offshore wind turbines and, thus, foregoing the assumed 25 percent per-kWh cost reduction from doubling turbine sizes to 13 megawatts?
The answer is displayed in the text box at right: $72 per ton of CO2 (equivalently, $80 per metric ton, or tonne; these figures decrease somewhat if we factor in separate methane fees such as the levy enacted as part of the Biden Inflation Reduction Act).
In other words, a $72/ton CO2 price would have given the offshore windpower industry the same profitability enhancement it thought it would reap from doubling its wind turbine megawatt capacities . . . but without the heavy blow rendered by the multiple Haliade-X blade failures.
Granted, there’s no actual link between instituting robust carbon-emissions pricing and easing up on the impulse to push technological advances faster and faster. Even with a $72/ton carbon price, offshore wind turbines would still be in a size race.
The point, rather, is to illustrate the economic power of carbon pricing. If a $72/ton carbon price could raise wind farm profitability by the same degree as a huge and perhaps premature push into bigger frontiers, imagine the leverage that robust carbon pricing could exert on every sphere of economic and physical activity.