Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

News How much for a nuclear power plant?

  1. Sep 14, 2010 #1
    http://counterpunch.com/wasserman09142010.html [Broken]
    Why Atomic Energy Can't Compete
    Is the Nuclear Renaissance Dead Yet?

    By HARVEY WASSERMAN
    Soaring costs at Vogtle, the US's one active new reactor project, have stuck Georgia ratepayers with $108 million in unplanned overcharges Currently calculated to cost a sure-to-soar $14.5 billion, the Vogtle project got $8.33 billion in federal loan guarantees from Obama in February. Citizen/taxpayer groups have since sued to see the details, which the administration is keeping secret. Georgia Power, which is building Vogtle, has already asked for another $1 billion rate increase. ​

    Is this a factually accurate accounting of the cost of this new nuclear plant?


    In the US, liability is capped at around $11 billion, even though the financial damage from a full-scale catastrophe could easily soar into the trillions. Minimum estimates from the 1986 Chernobyl disaster, which occurred in a remote, impoverished area, have exceeded $500 billion. By recent estimates the death toll is 985,000 and still counting. ​
     
    Last edited by a moderator: May 4, 2017
  2. jcsd
  3. Sep 15, 2010 #2

    CRGreathouse

    User Avatar
    Science Advisor
    Homework Helper

    I find slightly lower costs online, but it's basically accurate. Amortized over 30 years, that's 2.4 cents per kilowatt-hour (using a total production of 2.3 GWe). Adding in the cost of refined uranium at maybe 1.1 cents per kilowatt-hour, that still gives the project a lot of headroom for personnel, repairs, insurance, and profit, since current prices are perhaps 10-15 cents per kilowatt hour -- and prices are likely to rise, even after inflation.

    So expensive, yes, but not unreasonable.
     
  4. Sep 15, 2010 #3
    Does the author of your source realize that a "full-scale catastrophe" like Chernobyl is impossible in modern nuclear plants?

    Additionally, a 0.1 billion cost overrun is peanuts to a 14.5 billion dollar budget. They were off by less than 1%.
     
    Last edited by a moderator: May 4, 2017
  5. Sep 15, 2010 #4
    Nuclear power plant cost?
    Tree fiddy

    More BS from someone who don't know what they are talking about. Chernobyl comes up so often becuase it's pretty much the only horrendous incident in the history of nuclear power.

    Since the 60's and discounting Chernobyl (are we really going to base out view on a technology on a Soviet era constriction - lets face it they weren't exactly known for pushing safety), you can count the direct number of deaths caused in Nuclear reactor incidents on your fingers.

    People like the author of that article wind me up, it's like pointing towards the de Havilland Coment in the 50's and declaring all plane travel unsafe. Disredarding the umpteen thousand incident free flight-years.

    Also France has managed to pull off Nuclear power, with no problems at all.
     
  6. Sep 15, 2010 #5
    He's antinuke

    "Will this finally kill the much hyped "renaissance" of a Dark Age technology defined by quadruple failures in human health, global ecology, sound finance and increasingly shaky performance?"
     
  7. Sep 15, 2010 #6

    loseyourname

    User Avatar
    Staff Emeritus
    Gold Member

    I wonder how many people have been killed in hydroelectric dam disasters.
     
  8. Sep 15, 2010 #7

    Office_Shredder

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

  9. Sep 15, 2010 #8

    FlexGunship

    User Avatar
    Gold Member

    Not to be overly utilitarian, but if you're discussing the dangers of power generation (which is not the purpose of the thread) shouldn't you count it in terms of watt-hours per death. Obviously, higher would be better! More power with fewer deaths.

    Since I invented the unit, I'd like to call it the "toasty" (symbol is the Jesus fish, ichthys).


    -Wind is pretty bad at 6.66 teratoasties.
    -Rooftop solar is horrible at 2.27 teratoasties.
    -Hydro is okay if you ignore Banqiao (the Chernobyl of hydroelectric) at 10 teratoasties, but a crappy 0.71 teratoasties if you include it.
    -Nuclear has the best ratio at 25 teratoasties if you INCLUDE Chernobyl. If you don't include Chernobyl then it has a rating of 1875 teratoasties. That's 1.875 petatoasties!!!! (That number includes a single death that was attributed to radiological exposure of a plant worker. There is still debate over that.)

    For comparison, coal is only 0.006 teratoasties, and oil is 0.028 teratoasties.

    Banqiao was responsible for 26,000 deaths directly, and 150,000 from famine and disease after. Chernobyl was responsible for 56 deaths directly and 19 more later were attributed to it. I vote we stop talking about Chernobyl entirely, forever, in the context of nuclear safety. It essentially works out to a rounding error for coal or oil.

    EDIT: source: http://nextbigfuture.com/2008/03/deaths-per-twh-for-all-energy-sources.html
     
    Last edited: Sep 15, 2010
  10. Sep 15, 2010 #9

    Office_Shredder

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    You say potato, I say petoastie

    So we've determined that nuclear power is not as evil and not as expensive as the OP makes it out to sound. Is there anything left to talk about?
     
  11. Sep 15, 2010 #10

    CRGreathouse

    User Avatar
    Science Advisor
    Homework Helper

    Well, there's the issue of liability. As a free-market enthusiast, I agree with the OP on this issue: let the nuclear plants cover their costs. We don't want another BP Gulf spill...

    If private insurers won't cover them, in a pinch, the US government can sell a policy -- but I'd prefer that it be provided by private insurance companies (or even directly by reinsurers).
     
  12. Sep 15, 2010 #11

    Gokul43201

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    There seem to be some variances that need resolving...

    ... like the difference between 56 (or 75) and 985,000. I have no idea where the larger number comes from, but that difference can take 25 teratoasties and shrink it down to a mere 2 gigatoasties.
     
  13. Sep 15, 2010 #12

    CRGreathouse

    User Avatar
    Science Advisor
    Homework Helper

    FWIW, Wikipedia (citing [1]) claims as many as 4000 deaths once indirect cancer deaths are included, but doesn't even total as many as 75 direct deaths.

    This would give a figure of 0.47 TT, putting it below the other renewables but still well above coal and oil. Of course the non-Chernobyl number makes more sense to me, considering that even at the time that style of reactor was considered unsafe and wasn't really used anywhere but the USSR; they're certainly not being proposed today.

    [1] Elisabeth Rosenthal (International Herald Tribune) (6 September 2005). "Experts Find Reduced Effects of Chernobyl". New York Times. Retrieved 11 September 2010.
     
  14. Sep 15, 2010 #13

    Ivan Seeking

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    What is the worst case scenario given the assumption that terrorists take control of a reactor, who have all of the equipment, training, and knowledge needed to cause the most destructive event possible, using the number of deaths as a metric?

    When we talk about safety, we have to include the potential for damage if someone is out to defeat the system.

    There is also talk about limiting the size of dams. The idea that the failure of any constructed system could cause the death of millions, is called into question wrt more than just nuclear power. Frankly, I tend to think the Chinese are taking a big chance with the Three-Gorges Dam.
     
    Last edited: Sep 15, 2010
  15. Sep 15, 2010 #14
    I'm no nuclear engineer, but I'm under the impression that modern reactor designs have built-in fail-safes that simply cannot be overridden by human operators.

    http://en.wikipedia.org/wiki/Passive_nuclear_safety has a list of such fail-safes, but I admit I lack the nuclear engineering knowledge to understand much of it.
     
  16. Sep 15, 2010 #15

    CRGreathouse

    User Avatar
    Science Advisor
    Homework Helper

    Surely "worst-case" is the wrong metric to use here. In that case, it could be tens of billions for just about any scenario: the Earth's population grows, then experiences total existence failure.

    If we (reasonably) want to exclude the Banqiao Dam incident, caused by a '1 in 2000 year' flood, maybe we should consider events which have a 1/1000 chance of happening in a given year.

    This doesn't detract from your suggestion -- even with the current number of nuclear power plants I could see 1 or 2 such events happening in a dozen centuries, and more as the world moves away from fossil fuels. But I thought it important to draw this distinction early in the discussion.
     
  17. Sep 15, 2010 #16

    CRGreathouse

    User Avatar
    Science Advisor
    Homework Helper

    That's essentially right. (I don't do nuclear power, but I *do* work in radiation safety... if that counts for anything.)

    But that's more like 'resistance to meltdown' and less like 'resistance to terrorists'.
     
  18. Sep 15, 2010 #17

    Gokul43201

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Does it make sense to not include Chernobyl fatality numbers but still count pre-Chernobyl cumulative TWH? Also, accidents tend to be stochastic. It can't be good science to simply exclude specific accidents (from an already small sample) on the grounds that those particular accidents can no longer occur in modern systems.
     
  19. Sep 15, 2010 #18

    russ_watters

    User Avatar

    Staff: Mentor

    Based on that description, I'd say the "worst case" would be that they brought with them a 10 megaton nuclear bomb and detonated it at the nuclear plant. That's about the "most destructive event possible" by humans today.

    Not sure if it is very realistic to consider such a possiblity, though.

    Realistically, if they hijack the plant for a few days and have the expertise, I suppose the most immediate risk is of a major power failure if they do it in the summer. Long term, they could probably cause enough damage to necessitate closing the plant, which would cost several dozen people their jobs and badly hurt the shareholders of the company that owns the plant.

    [edit] Backup for the above opinion:
    http://www.world-nuclear.org/info/inf06.html

    In other words, barring a meterorite strike or terrorists trucking-in a rediculously large quantity of explosives (that may not be enough: you may actually need a nuclear bomb), TMI represents about the worst possible failure of a western nuclear reactor.

    [edit] I suppose there is another possibility: they could steal the fuel and truck it into the nearest city along with a conventional bomb. That would require holding off the Marines for a few days, though, which is pretty unlikely.
     
    Last edited: Sep 15, 2010
  20. Sep 15, 2010 #19

    CRGreathouse

    User Avatar
    Science Advisor
    Homework Helper

    Certainly I wouldn't want to count the production from any RBMK plant.

    Agreed. When I wrote 1/1000 chance per year, I was specifically envisioning that as lambda = 0.001 in a Poisson distribution.

    On one hand, I agree that naive extrapolation from a small sample is a bad idea; in particular, in this case, it inflates the "TT" rating.* But on the other hand, that Soviet design was known to be unsafe even at the time, and there's no expectation that more of that style will ever be built. So actually yes, I would say that good science allows -- even requires -- excluding that design.

    * I have a good example in mind, where a group of extremely low-risk drivers (?) were insured [with very low premiums] under the terrible assumption that their lack of accidents were due to their skill, where actually chance played greater part. Unfortunately I can't think of the name of the group! Our example doesn't have the same selection bias, but it's still similar in concept.
     
  21. Sep 15, 2010 #20

    CRGreathouse

    User Avatar
    Science Advisor
    Homework Helper

    This might cause thousands of deaths if there's a heat wave -- but of course this risk applies to all plants, not just nuke plants. (They're more vulnerable in that they tend to provide more power and thus service more people, but less vulnerable in that their security tends to be tighter. I'll call it a wash unless someone wants to crunch numbers here.)

    Worst-case loss in that scenario: perhaps $10 to $20 billion. At $10 million each, that could cost up to 2000 lives, in the sense that the money (which will eventually come from somewhere) could have been used to save roughly that many lives.

    I think this is what Ivan was referring to.* It would be interesting to do some Fermi estimates on the chances and potential damage caused. I would tend to think of that as major property damage (semi-permanent evacuation of a whole city!) but unlikely to actually kill many people. But I freely admit that's entirely speculation. Thoughts?

    * He may also have been referring to the possibility that they would create a nuclear weapon from the fuel. This isn't a risk today. I actually think this could become a major issue in the future, but not during my lifetime: the technology requires is too difficult and tightly-controlled. If a terrorist gets a nuclear weapon in the (not-even-that) near future it will be stolen (or bought), not created by the terrorist group.
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook