FOR THE WIN'

Jan 12
6 min read

I keep hearing folks in Europe and the United States dunking on wind power, claiming it has been a total failure, and offering this as strong evidence of the lie that always was alternative/renewable/green energy. Of course, in the current atmosphere, the alternative commonly presented is nuclear. Great. That gives us something relatively simple to compare to see if there's any truth to that.

The most recent nuclear reactors built in the US were Units 3 and 4 at the Vogtle Electric Generating Plant in Georgia. The cost for both units was $51 billion and they say the units took more than a decade to construct, with construction beginning in 2013 and commercial operation arriving around mid 2024. Now scheduled to come online in 2031, a similar pair of reactors at Britain's Hinkley Point C are set to cost more than $85 billion and, if they produce power in 2031, they’ll’ve taken around 13 years to build.

HUGE CAVEAT:

All these timeline estimates are just half the process. What 100% of people want to know is “If we decided to build a nuclear plant today, what’s a reasonable timeline from now to energy production and revenue?” I’ve never heard anyone imply anything else and don’t understand any other rationale… So then, I notice the planning for the upgrade at Vogtle, for example, began in 2006 and the first site permit and work authorization landed in 2009. At that time, construction was scheduled to reach completion in 2016. So, they thought it would take seven years just to build and turn on and it looks to me like it took closer to 18 years to go from design and approval to juice in the system… Hinkley Point was announced in 2008, construction of enabling works for ground investigation began that year, development applications were submitted in 2011, the European Commission approved the project and initial construction began in 2014. The completion date has moved and moved again and now sits at 2031. So not anything like seven years “to build a new nuclear reactor”... If you go looking for timeline averages you find most of the builds in recent decades have taken place in China and this is where the industry gets its numbers from. As we all know, China’s production and construction capacity, ability to massage regulation and steamroll approvals all while having non-existent labour laws doesn’t translate to anywhere else on Earth. Regardless, Chinese reactors have typically taken only around 60 to 90 months, five to seven years, to build. By contrast, recent reactors in India (Kakrapar 4) and France (Flamanville 3) took longer than America’s Vogtle 3-4. They tell us “construction” at Vogtle took 123 months while Kakrapar was 158 months and Flamanville 204. So not close to 60 months or even seven years but approaching twice that (even when you remove half the time from concept to energy production). Reactors with still longer “construction” timelines, as in Britain, are also currently in the works and don’t figure in any of these build-time assessments. So, by any rational assessment, for most populations in most places these days, from deciding you want a reactor or two to having those delivered and turned on you’re probably looking at decades…

Okay. So how long did it take to build those largest and most difficult to install ocean wind turbines you see? Well, our friends in China dropped a 20MW wind turbine, the largest and most powerful ever built, in the sea between Fujian province and Taiwan a year ago. What we know is this typhoon-resistant turbine, that sits about 174 metres above sea level, set a record for its 20MW capacity and also its rotor length: three 147-metre-long blades that sweep an area of about 10 football fields. This increase in capacity means that when compared with the 16MW units in more conventional offshore wind farms, it would take 25% fewer units and dramatically reduce sea area to achieve, say, a 1GW network (what we typically see of modern nuclear plants). Though the cost of production was not disclosed (China!), industry experts suggest the unit cost was somewhere in the range of $40-70 million. And how long did “construction” take? The manufacturing of this new wind turbine began in October of 2024 and it was delivered and fully installed, at sea, in January of 2025. Apparently the installation phase itself took less than three weeks. So, a period of around three months total from blueprints to power.

Now, to compare. According to the International Energy Agency and World Nuclear Association, offshore wind production capacity (or “capacity factor”) is around 45-55% and the global average for nuclear sits at around 80-90% of their nameplate capacity. Meaning: your "1GW" nuclear plant actually produces 800MW or, if you're lucky 850MW. So, in terms of real energy output, you would need about 85-95 of these new generation offshore wind turbines to produce the same amount of energy as a 1GW nuclear plant. Okay. So, with all these numbers to work with, if you suddenly find yourself with an unpowered AI supercluster in need of 1GW of energy or you're a municipality seeking to swap out your coal-fired power for something else and you could only choose between wind and nuclear, which system would be less expensive and which would be quicker to get up and going?

If a 20MW wind turbine costs $40-70 million, 90 of those would cost $3.6-6.3 billion. The time to manufacture and install 90 wind turbines would be months, not the 23 years it looks like if you just multiply the above three month production time by 90. Why? China installed more than 10,000 units per year over the last decade, with 14,400 installed in 2024 alone, to a total of about 220,000 units. As for offshore turbines, they’ve installed roughly 7,000 units in just the last five years, with as much as 17GW arriving from 3,400 units in 2021 alone. For comparison, Europe installed almost 7GW of new wind power in the first half of 2025 and is set to add another 178GW of new wind power between 2025-2030. So, fabricating and installing 90 units to 1GW in a few months, rather than a decade or two, doesn’t seem crazy to me.

Then, attempting to compare apples to apples, the quickest Chinese nuclear power plant of 1GW capacity took 61 months or just over five years to build and cost around $4.1 billion. Of course, folks in Europe and North America seem unable to do the same in under a decade or costing less than tens of billions as of late. Industry reporting suggests a single large offshore turbine in Europe can be erected in two or three days and a medium-sized wind farm of 50MW is typically constructed in about six months, costing between $240 and $300 million depending on the site. Double that cost and timeline for a 1GW offshore wind farm? Not sure.

Regardless, as I see it, all of this makes the speed of wind turbine installation and the cost, even at sea and especially when you factor in the possibility of electrons-to-customers by month three or six, to be a far better investment. And this is if you ignore the rather dramatic discrepancy in safety profiles, which everyone does and one must do to construct nuclear...

Looking around further I find the good folks at MIT's Climate Portal disagree with me, in a sense. Someone recently asked them "How many wind turbines would it take to equal the energy output of one typical nuclear reactor?" Their analysis came back offering: "Nearly 800 of today’s average-sized, land-based wind turbines—or, put another way, roughly 8.5 million solar panels." They get to this number by assuming much smaller capacity land-based wind turbines (averaging just 3MW rather than the above 20MW) and a reduced production capacity of just 36% (rather than 50%). In this assessment they also seek to produce 900MW. These are different calculations looking at different things. Crab apples to kumquats.