# Physical Insights

An independent scientist’s observations on society, technology, energy, science and the environment. “Modern science has been a voyage into the unknown, with a lesson in humility waiting at every stop. Many passengers would rather have stayed home.” – Carl Sagan

## Who’s not solving what fast enough?

How long does it take to construct a nuclear power reactor?

Westinghouse puts the construction time for the 1 GWe AP1000 nuclear reactor at 3 years; 36 months.  A more pessimistic estimate for current plants might be, perhaps, five years.

Let’s assume a nuclear power plant has a capacity factor of 80%.

That’s quite pessimistic – the industry in the US can and does often achieve capacity factors in the 90-100% range.

What’s the nameplate capacity of a wind turbine? Typically 2 MW. Let’s say that a wind turbine has a capacity factor of 33%.

(1000 * 80%) / (2 * 33%) = approx. 1200.

So, you need 1200 wind turbines to give the equivalent output as one typical 1000 MWe nuclear reactor.

How long does it take to build, transport, install and commission a wind turbine? Let’s say about three months. I heard that figure from an anti-nuclear-power pro-wind person once.
So, say it takes about three years to build a nuclear power plant. To build the equivalent number of wind turbines, it would be expected to take three hundred years.

Written by Luke Weston

January 17, 2008 at 12:20 pm

### 5 Responses

1. Obviously, 1200 turbines wouldn’t be built serially. Let’s see. Perhaps we should convert to units of person*time like man-years. Just guessing, but say it takes 1000 FTEs to build the nuclear plant and 10 FTEs to build each windmill.
1 plant * 1000 persons * 3 yr = 3000 man-years for the nuc.
1200 windmills * 10 persons * 1/4 year = 3000 man-years for the windmills.

I’ll take the nuclear plant because it’ll take up a lot less room than the 1200 windmills. Ever been to Tehachapee, California, USA or Livermore, California? What a blight compared to San Onofre Beach.

Nice to see you back. I really like your essays.

Reese

January 19, 2008 at 7:49 pm

2. I agree.

You don’t have to build the wind turbines serially, of course – but you don’t need to build nuclear units serially, either.

enochthered

January 20, 2008 at 4:05 am

3. The real important difference is not how long it will take to build the windmills, but rather the time it will take to update the rest of the grid infrastructure to accept that amount of intermittent generation. Since the answer to this question is somewhere between “God only knows” and “forever”, I think the windmills have an insurmountable disadvantage compared to the plug-and-play nuclear plants.

Sovietologist

January 20, 2008 at 7:57 am

4. Thanks, Reese, for the much-needed point. It is important to note, however, that when windmills “take up space”, less than 1% of the space over which the windmills are dispersed is actually claimed for foundations and access roads, etc. When this is done over farmland, the other 99% of the acreage is still usable for farming.

It is possible that this could still cause problems if windmills are put in places that aren’t farmland, but it’s not a simple issue of them “taking up space”.

gorbachev

February 29, 2008 at 5:55 pm

5. enochthered, you don’t have to build nuclear plants serial, but any single plant must be built once at a time. I don’t think your point stands.

gorbachev

February 29, 2008 at 5:57 pm