Half a croissant, on a plate, with a sign in front of it saying '50c'
h a l f b a k e r y
Left for Bread

idea: add, search, annotate, link, view, overview, recent, by name, random

meta: news, help, about, links, report a problem

account: browse anonymously, or get an account and write.



Fail-Safer Fission Reactor

Secondary heat-removal if the primary fails
  [vote for,

First, a bunch of background info:
A nuclear-fission reactor is a device that needs to maintain a somewhat delicate balance in order to produce power at a chosen specific rate. The key thing to control is called a "chain reaction". Each fuel-atom that fissions can potentially cause, on the average, up to two-and-a-half other fuel-atoms to fission. And in turn those fuel-atoms can cause yet more fuel-atoms to fission, and so on.

In a device designed to explode nuclearly, we want to maximize the quantity of fissions that each fissioning atom causes, and we want the rate of fissions to increase as quickly as possible. In a power reactor, however, we want to allow fissions to cause greater numbers of fissions ONLY when turning the thing on, and SLOWLY ramping up the rate energy is released from near-zero to the designed power-rating of the reactor.

At the designed power rating, the reactor needs to ensure that each fissioning atom causes, as exactly as possible, only one other fuel-atom to fission, thereby maintaining the chain reaction at a constant power-production rate.

Therefore all fission-power reactors have a group of things called "control rods", which are able to interfere with the chain reaction. The reactor core contains a regular pattern of control rods mixed with "fuel rods". The control rods are gradually pulled out from among the "fuel rods" to allow a chain reaction to begin, among the fuel rods. The control rods are often pulled out individually and by different amounts.

It is a peculiarity of just about EVERY large electric-power plant, whether of nuclear or other type, that it needs a source of external electric power in order to start working. On the rare occasions when some section of a national power grid fails, and a widespread blackout occurs, one of the reasons it can take days to restore power is that, usually, each power plant inside the affected zone needs to receive some external electric power in order to become turned on again. (And, sections of the grid need to be temporarily isolated, as explained below.)

Each type of power plant uses that external electric power in different ways. A hydroelectric facility, for example, may use external electric power to control the valves that allow water from the dam to reach the power turbines.

I don't know that the valves automatically close in the event of a power failure, so possibly hydropower is a little less susceptible to the generic problem just described. On the other hand, no single power plant can run the entire power grid by itself (its generators would overload), which means that usually each power plant in the grid MUST shut down, if only to protect itself.

Power-plant power can always be used to close the valves. But after that, then external power would be needed to open them again! After the local section of the grid is isolated, an external feed can allow the dam to once again begin to power its local section.

In a nuclear-fission plant, external power is used for such things as running the cooling pumps for the reactor core, and for moving the control rods. Once power starts being generated, of course, the external power is not needed. But if the internal power generation fails....

Recently in Japan, as most of us already know, a fission-power plant at a location named "Fukushima" lost power as a result of one of the largest earthquakes in recorded history. When such an event happens, the power-plant operators don't immediately know the extent of the damage. It is naturally DESIRABLE to restore power as quickly as possible.

"External power" does not necessarily automatically mean power from a distant source. The Fukushima facility had an auxiliary emergency power plant (fossil-fuel power), which naturally could be used for such purposes as initially getting the main fission reactors up and running --or to take over the powering of the reactor-coolant pumps. For a short time, until the fossil fuel tank was exhausted.

Well, the damage caused by the earthquake (and the tsunami that followed it) was rather more extensive than was initially understood to be the case, by the power plant operators. According to the "Pro-Nuclear" link below, all the reactors correctly shut down their chain reactions (control rods were pushed back into the reactor cores) in the aftermath of the quake. But the coolant pumps were not all working!

There exists a kind of "thermal inertia" which must be dealt with, even after a fission chain reaction stops. That's because not all of the energy of a fission reactor comes from nuclear fission; quite a lot also comes from the radioactive decay of the various atomic nuclei created by fission. So, that decay-heat still needs to be carried away from a "shut-down" reactor core. How much heat?

It happens that a "typical" large power plant these days is rated at approximately 1 Gigawatt of electricity production, 1000 Megawatts, enough energy to power 10 million old-fashioned 100-watt incandescent light bulbs. It also happens that a large modern heat-creating power plant is able to generate electricity from that heat at a rate of about 50% efficiency. This means that the amount of heat-energy being produced must actually be 2 Gigawatts, for 1 Gigawatt of electric power to be produced.

The Fukushima facility had 6 operational reactors, with 2 more planned. 3 of the operational reactors were down for maintenance at the time of the earthquake; the other 3 were running at their design capacity (see "Disaster Details" link for the individual power ratings). Even if only 1% of the total heat-production of the reactors was from radioactive-decay products, that's still roughly 1 Megawatt of heat (your oven is rated at maybe 4000 watts) that MUST be removed as fast as it is produced, lest it cause various bad things to happen....

The damaged systems were unable to cope with this heat-production, which has since increased the scope of the crisis at the Fukushima power plant (the related linked articles may need some updating, as both were written in March).

This Obviously Means More Types Of Backup Heat Removal Are Needed. Which brings us to the present Idea.

There is a "passive" type of heat-removal system known as a "heat pipe". See link. In theory, multiple large-diameter heat pipes can carry an enormous amount of heat from Point A to Point B. And heat pipes don't need external power to do their jobs!

So suppose each Point A, of several heat pipes, is a location NEAR, but not quite adjacent to, a nuclear reactor core. Each associated Point B could be at any nearby body of water (build several large artificial cooling ponds, if necessary). The reason we don't want each Point A to be right next to the reactor core is that we want this heat-conduction system to be used only in an emergency; we don't want heat escaping along that path when we are trying to produce electric power!

Note that the heat-pipes themselves should be made of a somewhat flexible material, so that they don't break (they would bend) if a large earthquake sends shock waves through the region.

If an emergency shut-down happens, now we do want each Point A to somehow become adjacent to a reactor core. I'm envisioning a kind of rotate-able "plug" of solid material located in-between the reactor core and each Point A. Each rotate-able plug would consist of two completely different materials.

At least half the plug would be an excellent heat-insulator, for when the reactor is producing power normally. When an emergency shut-down happens, and while the control rods are pushed into the reactor core, each plug is physically rotated so that instead of an excellent heat-insulator being between the reactor core and each Point A, the near end of one heat pipe, now an excellent heat-conductor is in place. The very best solid heat conductor is diamond, so that's the logical choice.

Each piece of solid diamond (it will have to be synthetic "grown" diamond, of course!) now conducts large amounts of heat from the reactor core to the Point A end of one of the heat pipes. Each heat pipe (remember, multiple heat pipes have been specified) now carries heat to Point B and the cooling water. Even if the normal cooling system fails, and the power plant goes dark, this system should prevent the sort of increasing emergency that happened at Fukushima.

Vernon, Apr 14 2011

Pro-Nuclear stuff, even after the crisis http://www.greentec...nuclear-imperative/
The basic need to stop burning fossil fuels has not changed. For one reason, try to find the May 2011 issue of "Popular Science" --an article describes how increasing CO2 levels is affecting the oceans, badly. [Vernon, Apr 14 2011]

Disaster Details http://newmediajour...s/indx.php/item/822
As mentioned in the main text. Likely needs to be updated with more recent information. [Vernon, Apr 14 2011]

Heat Pipes http://en.wikipedia.org/wiki/Heat_pipe
Operational details. [Vernon, Apr 14 2011]

"Popular Science" article http://www.popsci.c...011-how-save-oceans
As mentioned in the first link above [Vernon, Oct 27 2011]

Pebble Bed Reactor http://web.mit.edu/pebble-bed/
Why aren't we building these things? [Alterother, Oct 27 2011]

Reactor To Go I_20Got_20Your_20Nu...ctor_20Right_20Here
[theircompetitor, Oct 27 2011]


       Reads //First, ...// ... suspects ... scrolls ... suspects more ... scrolls ... confirmed.
spidermother, Apr 14 2011

       That it's a Vernon idea?
neelandan, Apr 14 2011

spidermother, Apr 14 2011

       Idea: use heat pipes to prevent nuclear disaster.
daseva, Apr 14 2011

       I think the idea is great except for the part about needing enormous diamonds.
That's probably a deal-breaker until we've got a way of cheaply manufacturing enormous diamonds.

       To be honest, I don't know why they can't use the residual heat generation to generate electricity.
Loris, Oct 27 2011

       Isn't this problem completely avoided by the German 'pebble bed' reactor <link>? Or am I hopelessly out of my depth again?
Alterother, Oct 27 2011

       Yes, but the problem with the pebble-bed design is that it's conceptually simple and doesn't use huge diamonds.
MaxwellBuchanan, Oct 27 2011

       Ah. I should have spotted that immediately.   

       Excellent work, as usual, [Vernon]. This is exactly the sort of thing we all love to expect from you.
Alterother, Oct 27 2011

       I suggest a new unit of measurement - the Vernon, which is a unit of measurement based on how many full screens it takes to display an idea from top to bottom.   

       Due to hardware and software differences, it would be a flexible measurement, but this one is about a six on my work computer.
normzone, Oct 27 2011

       //First, a bunch of background info://
...16 paragraphs...
//Which brings us to the present Idea.//

       This is quite good. The lengthy introductory section is clearly demarcated. The content seems OK, too. [+]
mouseposture, Oct 27 2011

       The Vn variability needs to account for tablets as well. On my iPad, this halfbake is exactly 6 Vn in landscape view, but only about 2.2 Vn in page view.
Alterother, Oct 27 2011

       // exactly 6 Vn in landscape view, but only about 2.2 Vn in page view.//   

       I make it 3.44 hours.   

       I am resisting the temptation (lest it bulldoze the idea) to suggest some sort of competition to see who can summarize the meat of [vernon]'s ideas in the form of a limerick.
MaxwellBuchanan, Oct 27 2011

       I call paralepsis!.   

       Heatpipes are great! (See the <link>).
And would solve a big problem, I think.
In a meltdown, some day,
they could move heat away
to some suitably distant heat sink.
mouseposture, Oct 27 2011

       I cannot possibly condone [mouseposture]'s annotation, even though it superbly contructed, well-rhymed, and could potentially save 3.433 hours.
MaxwellBuchanan, Oct 27 2011

       Furthermore, I would entreat halfbakers to resist the temptation to revisit all of [vernon]'s ideas simply in order to construct such time-saving limericks.
MaxwellBuchanan, Oct 27 2011

       Folks, I suggested solid diamond only because it is the best solid heat conductor. However, certain other solids are probably also workable. Remember that the hardest materials tend to conduct heat in the form of sound-wave vibrations, and the speed of sound in those substances increases with hardness. Sapphire might be hard enough. Silicon Carbide is likely hard enough. And both are routinely synthesized in quantity.
Vernon, Oct 28 2011

       I think having any additional engineering in close proximity to the core is a bad idea - it's a nasty place, and a very expensive place to do engineering in.   

       If you are going to use heatpipes, then why not use them more simply and put the complex engineering outside the core? Use them instead of the active cooling loop, to carry heat from the core to the generators. Then, in the event of an emergency, the switchover from generation to heat-dumping can be made at the far end of the heatpipes instead of at the reactor end - a much simpler and safer option.
MaxwellBuchanan, Oct 28 2011

       Perhaps this is the time to point out that thermo- syphon cooling of scrammed reactors is Baked, and has been for decades ?   

       There are a number of mitigating features that can be incorporated into reactor design briefs to prevent a recurrence of Fukushima, the most significant one probably being Do Not Build Your Reactor In An Area Vulnerable To Tectonic Activity.
8th of 7, Oct 28 2011

       //Perhaps this is the time to point out...//   

       Well, either now or shortly before [vernon] posted...
MaxwellBuchanan, Oct 28 2011

       [MaxwellBuchanan], can heat pipes handle the load of 2 gigawatts of heat when the reactor is running at full power? Remember heat pipes work semi-passively, without active pumping of the heat transfer fluid, and so, while I assume they can work well enough to carry residual heat (and the extra heat of radioactive decay) from a shut-down reactor core, I shy from assuming they can handle the full load.   

       Next, the engineering I had in mind is something that would be done when the reactor is first built. An ordinary pressurized-water reactor has a big steel pressure tank surrounded by lots of concrete. This Idea would have, embedded in that concrete, the rotatable plugs that would activate the emergency heat pipes. And of course all the near ends of the heat pipes would be embedded, too. Which means all this stuff simply gets placed before the concrete is poured. What's so difficult/hairy about that?   

       [8th of 7], this Idea is at least independently original with me. Can you post a link, please? Thanks!
Vernon, Oct 28 2011

       Well, you've got to have *some* system in there to carry off 2GW of power, as well as the lower power after shutdown. Makes far more sense to build a heatpipe system which can cope with the former as well as the latter, instead of having two systems.   

       Also, you avoid the need for tonnes of synthetic diamond and rotatey things inside the reactor core.   

       Frankly, though, all accidents and near-misses in reactors that have happened to date would have been avoided in current reactor designs, some of which are already in operation.   

       The heat-pipe idea is ingenious, but it's a bit like proposing a new and improved method for stopping steam-engine boilers from exploding.
MaxwellBuchanan, Oct 28 2011

       //Do Not Build Your Reactor In An Area Vulnerable To Tectonic Activity.//   

       So basically what you're saying is that the Japanese shouldn't bother with electricity, then?   

       Because the reason they have so many nuclear power stations is that they don't have easier options - fossil fuels not being prevalant close to the edges of Tectonic plates.
Loris, Oct 28 2011

              //Do Not Build Your Reactor In An Area Vulnerable To Tectonic Activity.//      

       So... Earth is right out, then?
Alterother, Oct 28 2011

       //fossil fuels not being prevalant close to the edges of Tectonic plates// Wait, what about Arabia, California, Mexico, Venezuela ... ?
mouseposture, Oct 28 2011

       //Wait, what about Arabia, California, Mexico, Venezuela ... ?//   

       Exceptions that prove the rule?
Loris, Oct 29 2011

       "Notably rare exceptions" ;}   

       [8th] did say //vulnerable to// not //close to// The Fukushima reactors might, for example, have been built on higher ground, no?
mouseposture, Oct 29 2011

       //The Fukushima reactors might, for example, have been built on higher ground, no?//   

       They might have been, but then they might have issues with cooling under normal operation.   

       The designers were concerned about tsunamis - they built a sea wall against them. But it wasn't tall enough and was overtopped.   

       Even in spite of that, the only fatality was a crane operator - and noone died or is likely to die from the radiation released. For a 40 year old design experiencing an event well outside its design parameters, that seems like pretty good going - and compares very favourably against, for example, the nearby oil refineries.
Loris, Oct 29 2011


back: main index

business  computer  culture  fashion  food  halfbakery  home  other  product  public  science  sport  vehicle