Wednesday, October 21, 2009

Stewart Brand and Amory Lovins Debate about Nuclear on NPR's OnPoint

Tom Ashbrook from NPR's OnPoint got the two to cordially hash out their opposing views on nuclear. Though the conversation lasted for about 12 minutes, not much was actually debated. I guess a good debate is what the blogosphere is for.

So far I haven't seen much praise for Lovins' latest piece, in fact it looks like in the comments section at Grist, his supporters were rather thin. Rod Adams continues to take Lovins to town, Brian Wang at Next Big Future had a lot to critique and Sovietologist piped in. We, of course, are generating our thoughts but are waiting a bit to see how the debate plays out. It's been spectacular to see the nuclear industry's supporters expose and rip up the Rocky Mountain Institute's latest junk science.

6 comments:

Anonymous said...

Amory Lovins asserts that nuclear power costs too much. Do the math yourself. You will find that it is not as expensive over its long life span.

Capital costs for a 1GW reactor may run 4 billion dollars overnight cost. Add on interest and the price is nearly double. Most industrial loans, unlike home loans, are for shorter time frames often less than 10 years. Investors don't go for longer term industrial notes.

In recent years our fleet of reactors have maintained a mean capacity of 91% which in round numbers amounts to 8000 hours/year. In one year one GW reactor puts out a million/kW each hour for the 8000 hours or 8 billion kWh. At ten cents/kWr that is 800 million dollars/year or 8 billion dollars in ten years. That is a 10 cent/kWh surcharge will buy and pay for the reactor one decade. Now the utility must pay other costs such the cost of maintaining the grid and O&E on the reactor of 1.76 cents/kWh plus line loss and the cost associated with doing business. Of course the shareholders expect a profit so perhaps during the ten year period when the capital costs are being recovered the rate may be 20 cent/kWh. The good thing is that after that ten year period the new reactor which is licensed for 60 years with option for an additional 20 year renewal will still have has 70 year life expectancy. The low O&E rate of 1.76 cent/kWh will easily beat out dirty coal. That's a real deal. Our society is not good at understanding investments that have an 80 lifespan. The price tag is not huge when the long period of service is factored in. If the utility owns four additional reactors that are paid for, the capital costs can be spread among all the customers so that the cost will be a 2 cent/kWh. This extra cost will be for just the ten years. The long term investment in nuclear power looks to be unbeatable.

Let’s compare nuclear power costs with wind generated power. Denmark, the poster child for wind power, has the highest cost for residential power rate in Europe, at 40c/kWh. The cost of Denmark’s wind capacity to Danish consumers is exacerbated by its inability to use so much surplus electricity. The claim that Denmark derives about 20% of its electricity from wind overstates matters. Being highly intermittent, wind power has recently (2006) met as little as 5% of Denmark’s annual electricity consumption with an average over the last five years of 9.7%. The balance is sold to neighboring countries at bargain basement prices.

Denmark has effectively become the World’s leader of distributed power. Since the Danish grid has been based on Amory Lovins’ distributive power concept, we can compare the overall performance of the distributive system, with the nuclear dominated system of France. In 2008 the emissions from nuclear powered France ran about 6.2 tons per person, while Danish CO2 emissions equaled 9.9 tons per person, over 50% more than France. Next door Sweden which gets 50% of its electricity from reactors ran an even lower at 5.6 tons of CO2 per person. Lovins' theories are apparently of little value in the fight against anthropogenic global warming. John Tjostem

Ioannes said...

Westinghouse just got slammed by the NRC on its AP1000 design:

http://www.nytimes.com/2009/10/16/science/earth/16nuke.html?_r=1

That contributes to delays in AP1000 design certification and to increased cost.

Joffan said...

Ioannes, that NYT piece was the usual misreporting of any minor event that is not wholly positive for nuclear power. Here's the first para of the corresponding NRC press release:
The Nuclear Regulatory Commission staff has informed Westinghouse that the company has not demonstrated that certain structural components of the revised AP1000 shield building can withstand design basis loads. An NRC letter to Westinghouse states that progress on the shield building review will require the company to provide modifications to the design, as well as testing that demonstrates the building will perform its intended safety function under design basis loads. The staff will continue its review of the remainder of the AP1000 design certification amendment application.
and Westinghouse are hoping (and have great incentive) to keep on schedule despite this.

Now I don't particularly agree that the design needed to be changed anyway, but as it has been, the designers need to satisfy the regulators that it will work, and the regulators need to tell the designers if they're not convinced. That's all fine - regulation in action.

perdajz said...

Yes Joffan, I appreciate your comments, but it's still frustrating to think that this process is held up by just the thought of something that will never happen. In all LWR experience (10,000 Rx-yrs?), we've never seen anything like a DBA. And the AP-1000 is likely to have a much lower DBA frequency than that of a typical LWR in operation. Also, the AP-1000 was designed with a greater emphasis of severe accident phenomena. Core damage and containment failure frequencies will be fantastically low, and orders of magnitude lower than that of the current fleet, which is plenty low already.

While the NRC tries to figure out how many angels can dance on the head of a pin, those coal burners keep dumping their residue in the atmosphere.

Anonymous said...

"this process is held up by just the thought of something that will never happen."

So accident analysis should not be bothered with when reviewing new designs for certification?

Not even Westinghouse is claiming that.

"after that ten year period the new reactor which is licensed for 60 years with option for an additional 20 year renewal will still have has 70 year life expectancy."

US power reactors are licensed for an initial 40 year period, with an option to apply for a 20-year extension.

perdajz said...

Anonymous,

Please define for me an acceptable DBA annual frequency. You don't get to answer "zero". Is one in a million years OK?

Of course Westinghouse doesn't say this sort of thing in public. If I still worked in the nuclear power industry, I wouldn't say them either.

Westinghouse has already done plenty of accident analysis on AP1000. That's quite evident from the design, which includes many simple, but clever engineered safeguards. By any measure, the likelihood of a DBA is fantistically low. And remember the DBA is not vessel failure, which in turn is not containment failure. A DBA is just the start of a long series of events needed for any actual radionuclide to the releases.

There is no law stopping hundreds of coal plants from dumping waste into the air with a probability of 1.0. I don't see why we should hold up building the AP1000 because the NRC can't exactly determine the accident progression for a scenario that might happen once every million years.