[FoRK] Why Fukushima made me stop worrying and love nuclear power
sdw at lig.net
Thu Mar 24 02:04:41 PDT 2011
On 3/24/11 12:42 AM, Eugen Leitl wrote:
> We're running on 15 TW. A typical reactor is 1 GW. That's
That's 15 TW in 2008 for all energy used, period. And it went down 1.1% in 2009. Electricity is only 1% of that.  20% of the
US's electricity is already provided by 104 nuclear plants, so it only takes 400 more to be 100% nuclear. But hydroelectric is
already about on par with nuclear, so that would only mean 300 more plants.
Hard to tell how the electricity vs. oil mix will end up, but bio oil/diesel with a very efficient plant/algae source will likely be
interesting for at least non-fuel oil needs.
> 15000 reactors, arguably 20000 reactors by the time you're
> done. These need to be breeders, which are a much more difficult
> control space.
> Are you seriously suggesting that this planet can tolerate
> 20000 reactors without the equivalent of Chernobyl going off
> about every month or so, and making the bad old days of
> athmospheric nuclear tests looking like cakewalk?
As of 2009, there were 436 commercial land nuclear reactors already, producing almost a TW. (Plus 180 marine and 250 research
reactors.) Maybe you should be in your bunker already?
Electricity production should be as much hydroelectric, solar, wind, wave, and geothermal as possible with nuclear providing the
remaining. We should save coal for the post-apocalypse. With electric cars, perhaps some portion of oil usage can shift to nuclear
The only serious threat of a Chernobyl is the continued operation of obsolete reactors. This is probably mostly caused by A)
popular pressure that is blindly against any nuclear progress and B) lack of pressure to innovate and upgrade existing facilities to
modern standards. While B seems like the owner/operators fault, I think the real roadblock is A. With new plant research, design,
and production, old plants would have the benefit of consistent production rather than the current situation of more or less
one-offs that are carrying all of the overhead.
In any case, we're likely to see far more than 3, or 10, or 20x of current plants: It appears that the main solution to all of the
pressure against large plants is leading to building safe, small, self-contained units that are hermetically sealed and are returned
as modules for recycling:
> The HPM is sealed at the factory, sited underground, and eventually returned to the factory for waste and fuel disposition after a
> useful life of seven to ten years. The principle materials in the core are uranium nitride (UN) fuel, stainless steel as the
> structural material, lead-bismuth eutectic (LBE) as the coolant, quartz as the radial reflector, B4C rods and pellets for in-core
> reactivity control and shutdown. The LBE permits ambient pressure operation of core, eliminating pressure vessel requirements.
> The outer diameter of the entire reactor system, including the outer reflector and coolant downcomer, is limited to 1.5 m to be
> able to seal the reactor vessel system at the fabrication facility and transport it to the site in a conventional nuclear fuel
> shipping cask. The total mass of the reactor vessel with fuel and coolant is <20 metric tons.
These seem likely to be popular for insta-cities all over, or even autonomous buildings / campuses. Google data centers. Cruise
Or perhaps a cluster of them to make a large plant.
> Sorry, Dr. Strangelove, I'll be in my bunker.
Good documents for the US:
> existing U.S. plants are performing well. Nuclear power plants now operate at a 90 percent capacity factor, compared to 56 percent
> in 1980. Additionally and in contrast to oil and gas, nuclear fuel costs are low and relatively stable. Fuel costs now average
> less than one half cent per kilowatthour. This is well below the costs of major competing fossil fuels. Production costs for
> nuclear power, operation and maintenance plus fuel costs, are also low, averaging 1.8 cents per kilowatt-hour. This cost roughly
> matches coal and is significantly below the costs of operating a natural gas plant.
> The oldest reactors still operating in the United States were licensed in 1969.
> *56 countries operate a total of about 250 research reactors and a further 180 nuclear reactors power some 140 ships and submarines.*
> Today, the world produces as much electricity from nuclear energy as it did from all sources combined in 1960. Civil nuclear power
> can now boast over 14,000 reactor years of experience and supplies almost 14% of global electricity needs, from reactors in 30
> countries. In fact, many more than 30 countries use nuclear-generated power.
> About 140 ships are propelled by some 180 nuclear reactors and over 13,000 reactor-years of experience has been gained with marine
> Russia also operates a fleet of six large nuclear-powered icebreakers and a 62,000 tonne cargo ship which are more civil than
> military. It is also completing a floating nuclear power plant with two 40 MWe reactors for use in remote regions.
[Table showing that France gets 78% of its electricity from nuclear, etc.]
Not cross-checked yet, but seems correct:
> In 2007 Pennsylvania used <http://www.eia.doe.gov/cneaf/nuclear/page/at_a_glance/states/statespa.html> 152,000,000 MWhours of
> electricity. How much did our 5 nuclear plants contribute to our electricity needs? Convert the 76,000 million kwh nuclear
> generation by dividing by 1,000 to convert to MWhours and then multiply by 1,000,000 to convert from million MWhours to simple
> MWhours and the result is 76,000,000 MWhours. Almost half of the electricity needs of Pennsylvania were produced by 5 nuclear plants.
> How many wind turbines would we need to replace these 5 nuclear plants? Most of the turbines in Pennsylvania now are rated at 1.5
> MW, however the new ones constructed are rated at 2 MW. The wind turbine only produce electricity when the wind blows greater than
> 12 mph making their average output somewhere between 25-30%, we will use 27%. This means that a 2 MW tubines with a 27% output
> yields an average of .54 MWhours throughout the entire year. To find out how much electricity that is we must multiply by the
> hours in a year which is 8760. One 2 MW turbine makes about .54 MW x 8760 = 4,730 MWhours per year. We would need 76,000,000
> divided by 4,730 or 16,000 large turbines to replace our 5 nuclear plants.
[Plus, pointing out that wind is not constant, and doesn't tend to blow in PA at peak summer points.]
More information about the FoRK