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

How long can nuclear plants last without power? (CME hit)

  1. Jan 6, 2017 #1
    Last night I watched History channel 10-hours "Doomsday 10 Ways the World will End".. the others are far out like Gamma Ray Burst, Black Holes, Rogue Planet, Killer Asteroid, Nuclear War, etc. and I can only feel entertained but when I watched the Solar Storm part. I got alarmed. This has happened before in the 1800s. We are due for another one. And when the Coronal Mass Injection hit us. All power grids will be fried including transformers that will take 2 years to build. In less than 10 months. Half the world populations would die from starvations and riots. But what concerned me the most is its mentioning that nuclear power plants need constant electrical power to cool the nuclear rods. And their backup power only lasts a week. So without power, would all nuclear rods really melt down? The series mentioned 450 nuclear plants would face meltdown. If this really occurred. Would their still be areas anywhere in the world where the radioactive clouds won't reach? (for us lucky few who would survive) This is very horrifying.. what is your take on this and what realistic things must we do to avoid such global catastrophe? Without watching the show, I couldn't have been aware of it. but why are others not seemed to be worried?

    What follows is some basic about CME and the danger if in case you are not aware of it.


    http://www.popsci.com/science/article/2011-05/are-we-prepared-catastrophic-solar-storm

    One of the biggest disasters we face would begin about 18 hours after the sun spit out a 10-billion-ton ball of plasma--something it has done before and is sure to do again. When the ball, a charged cloud of particles called a coronal mass ejection (CME), struck the Earth, electrical currents would spike through the power grid. Transformers would be destroyed. Lights would go out. Food would spoil and--since the entire transportation system would also be shut down--go unrestocked.
    Within weeks, backup generators at nuclear power plants would have run down, and the electric pumps that supply water to cooling ponds, where radioactive spent fuel rods are stored, would shut off. Multiple meltdowns would ensue. "Imagine 30 Chernobyls across the U.S.," says electrical engineer John Kappenman, an expert on the grid's vulnerability to space weather.
     
  2. jcsd
  3. Jan 6, 2017 #2

    rbelli1

    User Avatar
    Gold Member

    It seems that these worst case scenarios only happen if everyone stands around doing nothing watching the world burn. Presumably some percentage of the population would get to work fixing the most important things like food distribution and dangerous sites like nuclear plants.

    BoB
     
  4. Jan 6, 2017 #3
    But the show mentioned the CME can wipe out all available transformers. And it would take 2 years to build a transformer and this can only be built in Germany and South Korea so in less than 2 months all nuclear power plants are already in meltdown. The shows said even now in the United States. Delivering a transformer from Germany just took too long because it needs special transportation. And remember we won't have any oil because all the pumps won't work.. but the problem is we can't build a replacement transfomers in less than 2 months.. so all 450 nuclear power plants are in meltdown already. So what is really our backup plans available at present should a CME hit tomorrow that fry all circuits much like simultaneous EMP from all directions in all areas of the world?
     
  5. Jan 6, 2017 #4

    rbelli1

    User Avatar
    Gold Member

    Lots of people will die (including yourself) if you don't get *something* working right now. Oh they only know how to make transformers on the other side of the planet so we all die.

    So the CME destroys all above ground oil too?

    I'm not saying things won't be bad. Just that in an emergency people will find ways to fix or at least mitigate the worst problems. A few smoking craters will be unfortunate in cases where those efforts fail but most of the planet will still be inhabitable. Also bad for the people in various conveyances that go boom without power or electronics.

    BoB
     
  6. Jan 6, 2017 #5

    The show mentioned all our foods were processed by electricity and in refrigerators.. so after all electrical network wiped out.. we would run out of foods in a month, then with no water and no transportation to import it (remember all airplanes have their electronics wipe out as well as all boats), then there would be mass riots where your home will be invaded by others to steal anything you have left. This will go on for months then after the 450 nuclear plants meltdown, you need to leave your home and run.. but where? This is serious. We can ignore other ways the world will end but the coronal mass injections is the most scary because it happens once in 100 years and we are long overdue so it can happen within this year or soon. Watch this so you will be aware of the gravity of the situation:

    http://www.history.com/shows/doomsday-10-ways-the-world-will-end/season-1/episode-4

    I need to confirm that all nuclear fuel rods need electrity to maintain? Without electricity, it can really meltdown? How true is this?
     
  7. Jan 7, 2017 #6

    Astronuc

    User Avatar

    Staff: Mentor

    A CME will not necessarily result in damage to nuclear power plants, since they could isolate from the grid. One describes a loss-of-offsite-power (LOOP) or station blackout (SBO) situation. NPPs have onsite diesel generators to provide necessary power.

    Since Fukushima Daiichi inundation by tsunami (and loss of EP and ECCS), the nuclear industry has had to address Beyond Design Basis Accidents (BDBA), which are now a new (extended) basis. We have an array (spectrum) of Anticipated Operational Occurrences (AOOs) and Postulated Accidents (PAs), which the plants are designed to handle, and the objective is to prevent these from becoming more severe.

    While much of the economic infrastructure depends on electricity supplied from the interconnected grid, many facilities have independent and on-site generation, so not all is doom and gloom. Unfortunately, shows like the one cited sensationalize catastrophe in order to hook the audience; it is entertainment rather than education.
     
  8. Jan 7, 2017 #7

    etudiant

    User Avatar
    Gold Member

    While not disputing any of this, I do think we underestimate the impact of a large solar flare.
    The Carrington event in 1859 induced enough power across the telegraph grid that operators could disconnect the battery and still communicate.
    I don't know how well our grid today would handle that, but suspect the solid state components would prove painfully vulnerable.
    Could widespread blackouts be averted in that case? I'd be skeptical, because afaik, the Carrington flare impacted the grid for many hours, rather than just producing a single surge.
    A worse case scenario would include destruction of most inverters and rectifiers in our electrical infrastructure. Nuclear plants are not required to have the kind of heroic EMP shielding that Air Force One receives, so they may well be at risk much as the OP suggests.
     
  9. Jan 7, 2017 #8
    But with all the power grids shut down from damaged transformers. There is no pump to get oil.. and no fuel for airplanes or boats to deliver them and much of the roads would be blocked by massive riots and civil unrests by starved public.. remember there is no food because no deliveries so it is difficult to deliver the oil to the 450 nuclear power plants.. Unless these plans have many tons of diesel or gas as standby?

    Is there no way to create nuclear plants that don't use fuel rods that depends on water to cool them? Maybe we shouldn't have created nuclear power plants knowing major CMI that can disable power grid hits earth once in 100 years and we are long overdue.
     
  10. Jan 7, 2017 #9

    Astronuc

    User Avatar

    Staff: Mentor

    No, the fission process produces heat, and an inherent by-product of the process is fission products, which also decay. It is the decay heat from fission products which must be removed by cooling.

    I believe power relay technology is adequate to protect the grid more so than the older mechanical relays of 1800s.
     
  11. Jan 7, 2017 #10
    In an "unmitigated" total station blackout, meaning a total loss of offsite power, no injection, no steam powered cooling pumps, nothing works, you would boil off your inventory and have core damage begin within 1-2 hours (less for BWRs, more for PWRs) for a generation 2 or 3 light water reactor. Obviously advanced passive plants like the AP1000 can go up to a week or more, and small modular reactors like NuScale should be able to stay cooled indefinitely due to their small size and low decay heat loads. Containment failure will likely happen 24-36 hours after the initiation of the event.

    The reality: Nuclear plants have emergency generators. Not all equipment everywhere would be lost. There's a lot of standby equipment, there is fuel oil in tanks for strategic reserves, there are engine powered standby generators at critical installations and military bases, and nuclear power plants get priority for these resources when it is necessary to ensure core cooling, especially during the early phases of an event where there is still a lot of decay heat. At Browns Ferry in 2011 when multiple large tornados caused a loss of offsite power, one of the Browns Ferry reactors had an emergency generator fail and the military air lifted replacement parts from across the country within a couple hours to allow them to rebuild that emergency generator and get it running again. So there is history of prioritizing the restoration of power to nuclear plants.
     
  12. Jan 7, 2017 #11
    For above ground tanks, you only need the pressure head in the tank to get fuel oil out.

    Small modular reactors like the NuScale reactor are intended to be air coolable before their water supplies are depleted. It is possible. And all nuclear plants have at least 7 days of fuel (assuming 102% output from your emergency generators). During a loss of power scenario, you don't use most of your emergency systems, so you use much less fuel than full load and will have over a week of fuel easily. Even more if the plant decides to down all but one engine and just run one at a time.
     
  13. Jan 7, 2017 #12
    Can the temporarity fuel and generators last up to 2 years? remember it is said only germany and south korea can manufacture the giant transformers that can power the grid but even these plants can get out of commission sooner or later the 450 nuclear plants in the world can suffer meltdown...
     
  14. Jan 8, 2017 #13

    etudiant

    User Avatar
    Gold Member

    Are there not separate failure modes involved?
    The transformers suffer because the swings in voltage from the flare burn out the wiring, which is very exact and very considerable in any large transformer.
    The smaller stuff dies because the stray voltages that get induced damage the now universally used solid state ignitions and fuel controllers.
    So there may be much more vulnerability than expected.
     
  15. Jan 8, 2017 #14

    rbelli1

    User Avatar
    Gold Member

    I pretty sure nuclear reactors can be shut down in well less than 2 years. I would assume that the on site fuel requirements allow for a graceful shutdown in emergency conditions.

    BoB
     
  16. Jan 8, 2017 #15

    rbelli1

    User Avatar
    Gold Member

    I might be able to believe that only Germany and South Korea have plants set up to produce those transformers. I refuse to believe that everyone would stand around dying waiting for Germany and South Korea to produce them. Transformers are not exactly the most complex technology. Manufacturing plants in many parts of the world could be geared up for transformer production.

    Also the existing transformers could be disassembled and repaired. That may not be the proper course of action if one of them dies but if the health and safety of the entire planet was in jeopardy we could make some exceptions.

    BoB
     
  17. Jan 8, 2017 #16

    Astronuc

    User Avatar

    Staff: Mentor

    Shutting down a reactor can occur in seconds, but the shutdown is not the concern. Rather, removal of decay heat from the core is the main concern following shutdown, and decay heat (from decay fission products) happens at its own natural pace.
     
  18. Jan 8, 2017 #17

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2016 Award

    Has anyone ever heard a nuts and bolts explanation of how the big transformers get wrecked? Protective relays disconnect them from the long wires first thing.
    Relay equipment is built to withstand surges aroiund 1.5kv, see IEEESWC.

    While a Carrington like event is way more plausible than all the word's computers stopping on 1/1/2000 or 12/31/2012

    i agree it's become a Tempest in the Teapot purposed to sell commercials.
     
  19. Jan 8, 2017 #18

    etudiant

    User Avatar
    Gold Member


    All I know is that a relatively minor solar storm fried a good part of the Quebec Hydro system in 1989.
    As a New Yorker, whose power is going to depend on Quebec Hydro even more after our Governor shuts down Indian Point, I take that seriously. I see the result, but don't have the details, although the issue afaik is that there is a hell of a lot of fine wiring inside a transformer which overheats and burns thanks to the induced currents. That stuff fries once and is beyond repair, if the pictures are any guide.
    There is a separate issue of small scale ignition systems and fuel pumps losing it because of EMP, afaik there are no standards and no actual testing is done.
    The USAF has a very impressive installation out at Kirtland Lake called Trestle, a wooden trellis that supports an all up B52 to allow EMP testing.
    No commercial product has ever been verified similarly afaik.
    So I'm skeptical about the quality of the preparations for a Carrington class event. Imho, they are about as credible as Europe's precautions in advance of another large tsunami such as the one that devastated Lisbon in 1740.
     
  20. Jan 8, 2017 #19
    Does anyone have a list of locations of all nuclear plants worldwide? If United States is more organized, then maybe only the nuclear plants in the US and Canada would be secured in the event of a total grid wiping CMI? Imagine worldwide it's like Syria civil war in many countries with the populations no food after months of the initial CMI and massive civil unrests. In case 150 nuclear meltdown occurs elsewhere except the United States. Would radioactive clouds hit the west? For long term plan. I'm planning of moving into the safest location because it's not the question of if CMI grid killer event will hit.. but when and how soon.
     
  21. Jan 8, 2017 #20
    Transformers and such will not be destroyed by a CME event if they are not in use at the time.
    These days we should usually have advance information of a major CME occurring that might impact Earth.
    So most parts of power grids could be turned off for an hour or two - freaky, but much better than a decade long reconstruction job.
     
    Last edited: Jan 8, 2017
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted



Similar Discussions: How long can nuclear plants last without power? (CME hit)
Loading...