Nuclear energy

[BattleMoose]

And just to keep our facts straight it could have been much worse.

Could you paint for us, a worst case scenario?

[HungryHobo]

bigger fires, fires hitting more important parts of the building, more problems with cooling, more material released.

fukushima was bad as far as accidents at a plant with an actual containment structure go but it wasn't an absolute worst case.

it's not hard to imagine things which could have made it worse.

[lutzj]

bigger fires, fires hitting more important parts of the building, more problems with cooling, more material released.

fukushima was bad as far as accidents at a plant with an actual containment structure go but it wasn't an absolute worst case.

it's not hard to imagine things which could have made it worse.

So, it could have been worse but only if the containment structure was blown up by Dr. Evil, or something else improbable (perhaps still worth considering when assessing risk)?

[morriswalters]

Greenpeace just loves arguments like that. I don't know that there is an equivalent, or that it's important. Of course the small local businessman who lost his business and his home because effectively his community disappeared may feel kind of bad. And when asked if he still supports nuclear power, he may not feel particularly cheery about the prospect. Now how valuable that square mile is has a lot to do with it. Suppose it had been New York. Indian Point is fairly close. Look at it's location. It's in Buchanan, New York. That's on the Hudson River. Upstream.

Your not going to win this argument(not with me, but the public) by fancy footwork. I don't trust the herd, but what they lack in knowledge they make up in numbers. Frighten them bad enough and nuclear power could go the way of the Passenger Pigeon. I hate to say this because it's not really fair, but some action, like the resignation of the board would be a nice symbolic gesture which might go a long way towards making people more accepting. And you can be glad this happened in Japan and not the US. They are a people to whom value cooperation and compliance as part of their culture. Here there would be much wailing and gnashing of teeth, lots of blood lust and lawyers out the ying yang.

In terms of how bad it could be, tell me the type of accident that will happen next.

[mosc]

It's the whole plane crash vs the car accident thing. You can tell people till you're blue in the face how much more likely it is they'll die driving to the airport than they will in the plane once they get there but they fundamentally won't believe you. The human brain doesn't do statistics, it does pattern recognition. It recognizes the pattern of safe car use because the car is used so much more frequently. The plane is used far far less frequently so any imagery shown involving a plane crash will form part of an incorrect mental pattern. It's a common enough phenomenon but one most people rarely divorce themselves from. Nuclear accidents are big, flashy, and interesting. People dieing of lung cancer are a statistic. You can show people all the statistics in the world that even with a Chernobyl sized accident every 100 years or so, Nuclear is still the safest power per MW we've ever come up with. They just won't believe you.

[EdgarJPublius]

literally the worst case scenario that could happen to a nuclear reactor(s) did, fukushima,[i] and no one died.

To be fair, other than the evacuation deaths, many plant workers seem to have received significant radiation doses.
http://en.wikipedia.org/wiki/Fukushima_Daiichi_nuclear_disaster#Casualties

while the doses were above regulated limits, these limits are actually quite conservative and the actual accumulative doses were rather small.


As far as the evacuation issue, I think it should be considered that fossil-fuel related incidents can be far worse. The Deepwater Horizon contaminated numerous coastal communities and may have done irreparable harm to food supplies that come from the gulf. The harm done by fracking is just now beginning to be understood, and has the potential to make affected communities uninhabitable.

Numerous chemical processes, including fuel refining, involves toxic chemicals that can leach into groundwater and affect communities for many years to come. Communities built on or near waste dumps or old refinery grounds may have to be evacuated decades after the original incident.

Just because 'radiation is invisible and can kill you' isn't really that big a deal, air pollution is invisible too and kills far more people, PCBs and heavy metals in your drinking water are invisible too

[morriswalters]

I know Zamfir thinks I'm nuts but people relate atomic energy to mushroom clouds and bombs. And they remember Japan, and the movie mythos created by movies like "Them".

[zmic]

literally the worst case scenario that could happen to a nuclear reactor(s) did, fukushima,[i] and no one died.

To be fair, other than the evacuation deaths, many plant workers seem to have received significant radiation doses.
http://en.wikipedia.org/wiki/Fukushima_Daiichi_nuclear_disaster#Casualties

while the doses were above regulated limits, these limits are actually quite conservative and the actual accumulative doses were rather small.

if doses above regulated limits are something to be pooh-poohed, why don't we simply raise the limits then?

[roflwaffle]

So, it could have been worse but only if the containment structure was blown up by Dr. Evil, or something else improbable (perhaps still worth considering when assessing risk)?

Yeah, something like an earthquake with a high PGA is potentially more damaging. Earthquakes as small as 6.X can have PGAs well in excess of what any nuclear plant has experienced so far, and if those come from previously unknown faults that are fairly close to the plant, eg Diablo Canyon in California, then there's the potential for a very costly situation. This probably won't be an issue for new power plants going forward, provided they take a good long look underground, but for existing plants it's a potential issue.

[Zamfir]

I just visited the construction site of the new reactor in Finland. It's huge. Most nuclear plants are slightly disappointing in real life. the reactor buildings in particular are easily overwhelmed by the turbine building. But this is one is seriously big, like a merger of a cathedral and a bunker. Expensive, though.

[Xeio]

if doses above regulated limits are something to be pooh-poohed, why don't we simply raise the limits then?

It's generally not a good plan to put the limit at the exact point people begin to die. Room between the legal limit when action is taken and when casualties start to pile up is a good thing, not bad.

[EdgarJPublius]

Well, it's a bit complicated. Some room between the 'limit' and the actual danger zone can be a good thing, but too much can lead to complacence, people ignoring the limits and other safety precautions. So you want the limit to walk that fine line between 'safe' and 'meaninglessly safe'.

The radiation dose limits have been criticized for being too low, but the general paranoia regarding anything nuclear probably means that the limits will stay low. Of course, that same paranoia also means it's less likely people will be careless or complacent about radiation doses.

[infernovia]

So this guy is backed up by greenpeace and other such organization, but is this a valid study of wind variability and the impact it has on the network? This is the study, btw, cited in the australia 100% renewable thing.

http://assets.wwf.org.uk/downloads/mana ... report.pdf

[elasto]

if doses above regulated limits are something to be pooh-poohed, why don't we simply raise the limits then?

We surely could, but there's such a paranoia about radiation it probably won't happen.

You've most likely come across this chart but some reading this thread might not have (heck, this thread is 12 pages, it's probably been posted before, but surely still worth posting again!)

[mosc]

So this guy is backed up by greenpeace and other such organization, but is this a valid study of wind variability and the impact it has on the network? This is the study, btw, cited in the australia 100% renewable thing.

http://assets.wwf.org.uk/downloads/mana ... report.pdf

I read a good bit of this. It's a good paper. Its easy to cherry pick things that were glossed over but I did find the discussion on the variability of wind power a little distorting. The power grid really is a "series of tubes". It can bottleneck, and there are consequences to such things. As far as removing these bottlenecks, people are increasingly against the construction of power lines in their area. The author smooths out all wind generation in the country of Denmark on a yearly basis. Using such a large geographic area and such a large time scale removes the discussion from the fundamental problem with wind (and solar too for that matter): what do you do with the unusual conditions? Power grids are designed to handle unusual conditions. We have a zero tolerance for large scale blackouts. They cost lives let alone the economic disruption. The 3% of the time when it's not windy across the entire country of Denmark is precisely the reason wind power is so problematic. The author points out that no power plant operates 100% of the time, which is of course true, but all practical solutions to the power problem must contain a huge amount of power we COULD make, but chose not to so that in case of emergency, it can VERY QUICKLY be utilized. This is what you gain by having a nuclear reactor running at half steam, or a coal power plant off but with a mountain of coal next to it. It's not the majority of the time that's the problem. The entire infrastructure, as well as whatever power solution we chose to implement, needs to focus on the worst 1%. This is inherent in any high-risk analysis and has nothing to do with power.

[roflwaffle]

I think that generally nuclear and coal are poor for load following. Nat gas is generally the most flexible for dispatchables, and even with nuclear power you need to have a backup when a plant goes down unexpectedly. The better question is, does N amount of nuclear with N/M amount of load following cost less/pollute less than N amount of wind with N/P (Where P<M, usually) amount of load following?

[HungryHobo]

I think that generally nuclear and coal are poor for load following. Nat gas is generally the most flexible for dispatchables, and even with nuclear power you need to have a backup when a plant goes down unexpectedly. The better question is, does N amount of nuclear with N/M amount of load following cost less/pollute less than N amount of wind with N/P (Where P<M, usually) amount of load following?

problem is: with nulcear or coal you do have to be able to handle a plant or possibly even 2 plants going down unexpectedly.
Every plant is an island.

with wind you have to be able to cope with every single turbines output potentially dropping to nothing since you have no control over when that happens.

[elbekko]

Wouldn't it be possible to have a nuclear plant follow load by using a similar principle to the variable vane turbo on the turbine, basically reducing the efficiency of the turbine so that less electricity is generated?
The biggest downside there is of course that while less electricity is being generated, the same amount of fuel is being used. And more cooling is probably needed because less of the steam's energy is dissipated into the turbine.

[HungryHobo]

And if you're gonna do that you might as well run it all the time and dump the cheap extra energy into anywhere it might be useful.

don't france do something like that selling cheap extra power to cern during the summer?

[mosc]

You know what would be great to do with cheap extra energy? Make Liquid Hydrogen for cars.

Just saying...

[Yakk]

And if you're gonna do that you might as well run it all the time and dump the cheap extra energy into anywhere it might be useful.

don't france do something like that selling cheap extra power to cern during the summer?

Less "selling", more "providing as their contribution", I believe.

[Soralin]

Wouldn't it be possible to have a nuclear plant follow load by using a similar principle to the variable vane turbo on the turbine, basically reducing the efficiency of the turbine so that less electricity is generated?
The biggest downside there is of course that while less electricity is being generated, the same amount of fuel is being used. And more cooling is probably needed because less of the steam's energy is dissipated into the turbine.

Or just design them such that you can change the power output of the reactor: http://en.wikipedia.org/wiki/Load_following_power_plant

Boiling water reactors

Boiling water reactors (BWR) and Advanced Boiling Water Reactors can use a combination of control rods and the speed of recirculation water flow to quickly reduce their power level down to under 60% of rated power, making them useful for overnight load-following. In markets such as Chicago, Illinois where half of the local utility's fleet is BWRs, it is common to load-follow (although less economic to do so).
Pressurized water reactors

Pressurized water reactors (PWR) use a chemical shim in the moderator/coolant (see nuclear reactor technology) to control power level, and so normally do not load follow. (In most PWRs, control rods are either fully withdrawn or fully inserted - variable control is difficult, partly due to the large bundle sizes.)

In France, however, nuclear power plants use load following. French PWRs use "grey" control rods, in order to replace chemical shim, without introducing a large perturbation of the power distribution. These plants have the capability to make power changes between 30% and 100% of rated power, with a slope of 5% of rated power per minute. Their licensing permits them to respond very quickly to the grid requirements.

It certainly can be done (and is being done), it's just that in places where nuclear isn't a large proportion of the power, they don't bother to. The cost of a nuclear plant is mostly in its construction rather than the fuel, so if you can reduce the power of any other plant first, that's the smart thing to do. If nuclear becomes a large proportion of the power supply, they'll simply use nuclear plants that can adjust their power output.

[mosc]

It's already done in France. Been done for decades.

[hawkinsssable]

So this guy is backed up by greenpeace and other such organization, but is this a valid study of wind variability and the impact it has on the network? This is the study, btw, cited in the australia 100% renewable thing.

http://assets.wwf.org.uk/downloads/mana ... report.pdf

I read a good bit of this. It's a good paper. Its easy to cherry pick things that were glossed over but I did find the discussion on the variability of wind power a little distorting. The power grid really is a "series of tubes". It can bottleneck, and there are consequences to such things. As far as removing these bottlenecks, people are increasingly against the construction of power lines in their area. The author smooths out all wind generation in the country of Denmark on a yearly basis. Using such a large geographic area and such a large time scale removes the discussion from the fundamental problem with wind (and solar too for that matter): what do you do with the unusual conditions? Power grids are designed to handle unusual conditions. We have a zero tolerance for large scale blackouts. They cost lives let alone the economic disruption. The 3% of the time when it's not windy across the entire country of Denmark is precisely the reason wind power is so problematic. The author points out that no power plant operates 100% of the time, which is of course true, but all practical solutions to the power problem must contain a huge amount of power we COULD make, but chose not to so that in case of emergency, it can VERY QUICKLY be utilized. This is what you gain by having a nuclear reactor running at half steam, or a coal power plant off but with a mountain of coal next to it. It's not the majority of the time that's the problem. The entire infrastructure, as well as whatever power solution we chose to implement, needs to focus on the worst 1%. This is inherent in any high-risk analysis and has nothing to do with power.

Any chance for some kind of academicish article on unusual conditions and wind, especially (if you happen to have one) for Australia? The few things I've turned up plugging some obvious terms into google scholar suggest variability is manageable, with a decent time frame to ramp up conventional or combustion power generation before these events hit. This could have something to do with google and confirmation bias, iunno. I have noticed lately that google scholar is especially annoyingly diligent in personalising its results for me (usually with articles I've already read.)

Also, a quick question on blackouts costing lives: where I live, they're irritatingly frequent even in winter because somebody thought putting the powerlines above ground in an area full of overhanging branches was a good idea. Everywhere where lives hang in the balance (hospitals, the nursing home I worked at, that kind of thing) have back-up power and extra generators in case something like this happens, and as far as I know nobody has died as a result. Is this different outside of Australia?

The better question is, does N amount of nuclear with N/M amount of load following cost less/pollute less than N amount of wind with N/P (Where P<M, usually) amount of load following?

That's the exact question I've been struggling to articulate!

[Griffin]

Everywhere where lives hang in the balance (hospitals, the nursing home I worked at, that kind of thing) have back-up power and extra generators in case something like this happens, and as far as I know nobody has died as a result. Is this different outside of Australia?

We had deaths from out recent major blackout in the NE US. Most places like you named have backup power, yes.... for a while. But if you get a long lasting blackout during a cold snap or a heat wave, people will (and do) die. (Though currently electric heat is extremely unpopular and not a super huge deal, but most of our discussion has been about getting off oil and natural gas, so I'd assume we'd have more of it as we move to renewables/nuclear)

[TranquilFury]

Everywhere where lives hang in the balance (hospitals, the nursing home I worked at, that kind of thing) have back-up power and extra generators in case something like this happens, and as far as I know nobody has died as a result. Is this different outside of Australia?

We had deaths from out recent major blackout in the NE US. Most places like you named have backup power, yes.... for a while. But if you get a long lasting blackout during a cold snap or a heat wave, people will (and do) die. (Though currently electric heat is extremely unpopular and not a super huge deal, but most of our discussion has been about getting off oil and natural gas, so I'd assume we'd have more of it as we move to renewables/nuclear)

Even gas fueled furnaces need electricity to power their blowers.

[morriswalters]

Blackouts have almost nothing to do with power generation, they will exist no matter the generating method.

[Griffin]

Blackouts are either a problem with power generation or power transmission. (since we were talking of larger grids with more distant power sources in the case of most of the alternatives being discussed, the second is also relevant.

Making your post is a great big wtf. Thanks for that wonderful contribution.

If you don't think blackouts are caused by something relevant to the issue at hand, what ARE they caused by? Do you think blackouts can't be caused by power generation problems? Since we were only discussing lacking power, in general, rather than blackouts in particular, do you also believe brownouts are irrelevant?

Essentially- what's your point?

[mosc]

I assume you didn't live in southern California in the late 90s...

[morriswalters]

In the US most blackouts aren't caused by a sudden failure of generation, though I could be wrong. The 2003 blackout was caused by heating in some power lines which I believed sagged due to heating in a high demand situation in Ohio, the blackout was a function of the system protecting itself as the load became unbalanced. Like most other things of this nature a combination of things made it worse. More commonly outages are weather related, caused by wind or ice. California has had rolling blackouts due to demand, but they were preplanned. The Texas wind farm that caused a power outage was a supply issue, there was capacity but it wasn't online, but the blackout lasted an only an hour or so. As an additional note, the Texas problem would not have happened, except that most of Texas, wisely in my opinion, is its own separate grid. Brownouts are not all that common and are generally of short duration.

Outages caused by severe weather are not blackouts in the classic sense. In most cases the grid never completely goes down, including the disruptions in the North East. They are service interruptions are caused by local damage, power lines down, breakers tripped and so on, last mile issues. Long lines aren't as sensitive to these type of disruptions because they don't in most cases have trees anywhere near them, I'm not sure about icing. Deaths in a large number of cases are a product of social isolation and carelessness. In most cases there were available options open to people who needed them.

Now no matter the method by which you generate power and where you generate it, you can't eliminate these types of problems, that is of course unless you generate it locally and bury every line. You might have heard of ideas about creating a power system composed of many small nuclear plants which may be safer then big ones. Or a fuel cell type that Google is using from Bloom Energy. Considering the natural gas that is going to appear from fracking I might buy one of Bloom energies devices myself, assuming that the price drops. Of course I am lucky I have natural gas. I also have a generator. Toodles

Mosc if you are talking to me then no I didn't live in California. But that was stupid at work, as it sometimes is there, not to mention some creative money making. It is also the reason I won't live in California ever. Best of luck to them, but no thanks. If anybody wants to site a nuclear plant next to my house, bring it on, I have no issues. We could build plants as fast as we could make them and it might never be enough. Oddly enough though this new influx of natural gas is evidently impacting coal fired plants, since evidently gas is cheaper overall then clean coal. As a slight funny the local utility is building a 50 foot tall fence so its customers won't have to see the ash left after all that coal is burnt.

[TranquilFury]

Outages caused by severe weather are not blackouts in the classic sense. In most cases the grid never completely goes down, including the disruptions in the North East. They are service interruptions are caused by local damage, power lines down, breakers tripped and so on, last mile issues. Long lines aren't as sensitive to these type of disruptions because they don't in most cases have trees anywhere near them, I'm not sure about icing. Deaths in a large number of cases are a product of social isolation and carelessness. In most cases there were available options open to people who needed them.

you don't need nearby trees for weather to down a power line:
http://imgur.com/6FdbH,uKSYL,pL3ld

[r3dman]

Wouldn't it be possible to have a nuclear plant follow load by using a similar principle to the variable vane turbo on the turbine, basically reducing the efficiency of the turbine so that less electricity is generated?
The biggest downside there is of course that while less electricity is being generated, the same amount of fuel is being used. And more cooling is probably needed because less of the steam's energy is dissipated into the turbine.

Not like a nuclear plant cares about fuel when the quantities of U needed are infininitely small when compared to the coal needed by a fossil-fuel power plant.

[BattleMoose]

Wouldn't it be possible to have a nuclear plant follow load by using a similar principle to the variable vane turbo on the turbine, basically reducing the efficiency of the turbine so that less electricity is generated?
The biggest downside there is of course that while less electricity is being generated, the same amount of fuel is being used. And more cooling is probably needed because less of the steam's energy is dissipated into the turbine.

Not like a nuclear plant cares about fuel when the quantities of U needed are infininitely small when compared to the coal needed by a fossil-fuel power plant.

Of course it would be better if there was a pumped storage scheme nearby to take advantage of essentially unwanted electricity, actually this helps ease some of the problems with wind anyway.

[Zamfir]

Wouldn't it be possible to have a nuclear plant follow load by using a similar principle to the variable vane turbo on the turbine, basically reducing the efficiency of the turbine so that less electricity is generated?
The biggest downside there is of course that while less electricity is being generated, the same amount of fuel is being used. And more cooling is probably needed because less of the steam's energy is dissipated into the turbine.

Not like a nuclear plant cares about fuel when the quantities of U needed are infininitely small when compared to the coal needed by a fossil-fuel power plant.

Nuclear plants care a lot about fuel costs, those are around a third of the lifetime costs of a plant. Partially the cost of mined natural uranium(about half of the fuel costs) , plus the costs of enrichment and fuel manufacturing, and disposal at the end.

Also, most reactors need to shut down to refuel, so more fuel use means more time without income.

[r3dman]

What i meant is that we dont have to care that much about "wasting" fuel when the natural reserves of U will last for hundreds of years, more than enough time to find alternatives. For a coal plant that IS a problem. About the disposal, i dont think that adds too much to the total cost when the High-Activity waste is stored within the plant.

[lutzj]

What i meant is that we dont have to care that much about "wasting" fuel when the natural reserves of U will last for hundreds of years, more than enough time to find alternatives.

Doesn't mean it isn't crazy expensive and difficult/risky to process. Nuclear plants still have to compete cost-wise to justify the massive investment that goes into them.

[Zamfir]

The context here is load following: if you cannot reduce the thermal power of your plant quickly, can you afford to run it at full thermal power but not produce the maximum amount of electricity?

A (new) coal plant needs about 300gram of coal per kWh of electricity, which costs about 2 to 3 cents. Depending on where you are. It's cheaper close to an easy coal field, and coal prices fluctuate quite a bit.

For that same kWh of electricity, a nuclear plant burns about 0.3 milligram of plant-ready UO2 fuel (it takes several milligram of natural uranium to make that). That costs about 0.5 to 0.7 cents (depending on the uranium price, and the contract with your fuel supplier). Spent fuel costs are in the order of 0.2 or 0.3 cent on top of that, including the money set aside for end disposal.

That sounds like small grub, but let's imagine you have a 1GW nuclear power plant. You decide to run it at 80% of electric power for 12 hours every day, while still producing 100% of thermal power. That's a waste of at least 5 million euro a year, in wasted fuel alone. Margins on electric power are razor-thin, that 5 million would be a noticable chunk of your profits.

So in practice, a nuclear plant will only do load-following if the reactor is designed to reduce thermal power together with electric power. That's feasible, but most reactors are not designed that way.

[mosc]

Of course it would be better if there was a pumped storage scheme nearby to take advantage of essentially unwanted electricity, actually this helps ease some of the problems with wind anyway.

Doesn't even really need to pump. It's like a self-refilling battery with a maximum capacity. You can choose when to use it, to a reasonable extent. At least over shorter periods of time. But hydro is a limited resource especially in local areas that lack major rivers. Most of what is possible is already tapped out.

[thoughtfully]

What i meant is that we dont have to care that much about "wasting" fuel when the natural reserves of U will last for hundreds of years, more than enough time to find alternatives. For a coal plant that IS a problem. About the disposal, i dont think that adds too much to the total cost when the High-Activity waste is stored within the plant.

We aren't running out of coal. There's more than enough to turn us into Venus down there. Ok, maybe not. Still, it'd last longer than Uranium, at least the way we're using it now. It's silly, really. The problem with nuclear waste is that we're wasting it. We discard something like ninety percent of the fissionable fuel because it's cheaper to dig more out of the ground than to reprocess it. I can see an energy starved future where the poor wretches on the fringes of society get to mine old waste depositories for this stuff. Madness.

Anyway, Thorium would be better. Imagine all this nonsense with Iran and North Korea blowing right away like so much fog in the morning sun.

[Zamfir]

Be careful with the terminology. Reprocessing alone will only get you something like 15% more power from the same amount of uranium. France does this on a large scale, it's not a great success but it works. Reprocessing together with breeders would give you 50 times more energy from the same amount of ore, but the difficult part are the breeders, not the reprocessing. A working breeder is fairly easy to build, but many people have as yet tried and failed to build an affordable and reliable breeder.

So at the moment we're not really throwing away 90% of the fissionable fuel. It's more like there is 99% of the fuel that looks very close to useful in theory, but turns out much farther away in practice.