Gen 3 nuclear reactors

[BrownTown]

So, since the US seems to be on a so called "nuclear renaissance" based on Genereration 3 and 3+ nuclear reactor designs I was wondnering about your views on the saftey of such designs. On one side everything seems great, just look at some of the AP1000 specs, (50 percent fewer valves, 83 percent less piping, 87 percent less control cable, 35 percent fewer pumps). Thats a whole lot less stuff for things to go wrong in. This of course should help a good deal with the economics in that the reactors are cheaper and simpler to operate. On the other hand, the saftey mechanisms are implimented using so called "passive" saftey. This basically means that natural forces are supposed to be making sure that the reactor can not melt down. The laws of physics seem a good deal more reliable than backup saftey systems that hardely ever get used, but maybe its just me, but I still feal a little uneasy knowing that the saftey is based on computer models of how physical objects should interact. And these models are supposed to ensure that enough cooling is available to remove excess heat, or to shutdown the reactor properly etc..

Also, if you have any knowledge of the designs of these reactor it would be interesting.

AP1000
ABWR
ESBWR
EPR
System 80+
any others?

EDIT: added links, anyone have system 80+ info, cant find it on ABBs site...

 

[jagec]

You're not putting nearly enough faith in computer models...they're a lot more widespread, and reliable, than you think. Pretty much all the research we do ends up with a computer model as the end result, which is based on fitting a curve to a number of data points taken at various conditions. Since we can't measure every single condition possible, a model is essential to describe the behavior of whatever it is we're investigating. Keep in mind that every real engineering feat today, whether putting a plane in the sky, or building a massive new chemical plant, or designing the latest chip, is based on models. When failures occur, it's almost always due to the failure of a component, a substutition or small redesign in the building process that wasn't fully thought-out, or other details that have nothing to do with errors in the model.

 

[gsellis]

I like REMHD. Recoil Enhanced Magneto HydroDynamics. Found out about that tech looking up MHD. MHD was being investigated in the 70's for power generation with biomass generation plants. MHD is designed to be significantly more efficient than just stream generation and use steam generation at the same time.

Paper on REMHD for space travel power

Big paper on Gen IV reactors Pg 91 - called Radiation Enhanced MHD here.

 

[pcy]

Hi,

Without knowing a great deal about this I'd have to say I'm pretty comfortable with it.

I share jacec's view of computer models - if the job is done thoroughly and matched with the available experimental data, they do tend work very well, and cover the ground far more completely than any set of practical experiments alone could.

But there is another reason why I'm not concerned. I think this is the wrong problem to worry about:

Despite Chernoble and 3 Mile Island, the nuclear power generating industry actually has a very good safety record, and recent research suggests that the effects of moderate exposure to radiation are much less severe (indeed it may possibly be beneficial) than previously supposed. The actual levels of death/illness cause by Chernoble (other than to those actually woking on site) are 10s of times lower than originally predicted.

It's not as if Oil and Gas are 100% safe, or produce no carbon emmisions, or have no other environmental impact.

However, I am seriously concerned about the storage of the waste products - highly radioactive and poisenous substances with decay times measured in millenia. I think that's where the focus should be.



Peter

 

[CycloWizard]

I am a big proponent of nuclear power and of passive safety principles in general. Such approaches have been used in other processing industries for decades but are only recently working their way into the nuclear area.

I dislike the usage of the term 'computer model' because it has some sort of mystical connotation, like data is fed into a computer and it magically spits out predictions based on nothing in particular. In reality, it's not like that at all. There are multiple ways models may be generated, and only one of them allows the computer itself to generate the model. Even in this case, the computer isn't really generating the model; the data itself generates the model based on well-known statistical principles (e.g. an automated analysis of variance). The much more common approach is that somewhere along the line, an engineer created a working model of the problem. This working model includes all of the assumptions about the problem, such as the problem geometry, constitutive models, and boundary conditions. The human then constructs a solution domain over which the problem should be solved. The computer software then numerically solves the model that it has been presented. Thus, the model and possible solutions are input by a human. In essence, you ask the computer a question, put some limits on the answers, and the computer spits out the answer using pure math. Finite element analysis, computational fluid dynamics, even Excel's solver are simply software implementations of these ideas to standardize the approach so that each person doesn't have to do the insane math required to solve systems of partial differential equations every time they want to know the answer to a question. So, if the 'computer model' fails, it is actually a result of user error.

 

[Mark R]

Originally posted by: BrownTown
The laws of physics seem a good deal more reliable than backup saftey systems that hardely ever get used, but maybe its just me, but I still feal a little uneasy knowing that the saftey is based on computer models of how physical objects should interact. And these models are supposed to ensure that enough cooling is available to remove excess heat, or to shutdown the reactor properly etc..

The AP600 (on which the AP1000 is based) has had it's passive safety features validated with scale models (including a full scale model of the main gravity cooling circuit).

The AP600 was an earlier design which was completed and signed off several years ago. However, it became clear towards the end of the design-phase that customers were unlikley to want a small plant (<1000 MW). So, the AP1000 was created by scaling up the AP600.

The main passive safety system in the AP1000 is passive cooling - this allows residual decay heat to be removed from the reactor in the event that the normal heatsink (the cold water feed to the steam generators) is unavailable. It works by simple convection (the same way as a typical domestic hot water system works).

The reactor lies at the bottom of the containment building. Above the reactor and steam generator systems, about half way up the containment building is a huge water tank. It contains a heat exchanger that is connected one of the reactor cooling circuits. In the event of an emergency, valves on this circuit are opened, and water will circulate by convection from the reactor to the heat exchanger and back.

The tank has sufficient water to absorb 1 hour of decay heat prior to boiling.

The 2nd trick is to deal with the boiling water. The inner containment building of the AP1000 is made of steel (1 3/4 inch thick plate). The advantage of steel is that it conducts heat. Boiling water in the heat removal tank evaporates, contacts the cold steel and condenses, where it falls back into the tank via a series of gutters. The size of the containment building is sufficient to allow cooling of the building via convection of air on the outside. (This has been verified by simulation and by scale model testing in wind tunnels).

However, in case a freak weather condition prevents this from ocurring (wind blowing in the wrong direction and somehow reducing the convection flow) the air convection system can be supplemented. A large roof tank on top of the containment building can be used to supply water sprays which can supplement the cooling. This tank is supposed to be large enough that supplemental cooling could be provided for 72 hours.

More info on AP1000

 

[BrownTown]

also interesting is that these reactors have breeding ratios of .7-.8, so they produce more plutonium then current light water reactors. That means you need less U235 than normal, so something like 15% less fuel should help on the economics side of things.

 

[jagec]

Originally posted by: Mark R
The 2nd trick is to deal with the boiling water. The inner containment building of the AP1000 is made of steel (1 3/4 inch thick plate). The advantage of steel is that it conducts heat.

Steel (esp. stainless) is actually a pretty lousy conductor of heat, as metals go...certainly better than concrete, though

 

[BrownTown]

yeah, which brings you to the point that most reactors have a layer of concrete around the steel which improves saftey, but the AP1000 does not have that extra layer in order to allow the steel to be exposed for cooling. Of course it does still have the concrete containment building, but its doesnt have quite as much protection from impacts (many would be thinking planes here) as some other designs.

As for the thermal conductivity of steel, yeah its not as good as many other metals, but pretty much all metals are good thermal conductors. Also, the heat produced after the reactor SRAMs is only a few percent of the rated power, so you not gonna have to deal with anything near the full 3400 MWt that is usually put out.

 

[Tiamat]

I did some studies on Nuclear reactors this past year for my Senior Chemical Engineering Design Problem. If you want, I can send you a copy of my paper. Just PM me. It is somewhat large, but I do not think it would be difficult to read through.

 

[inveterate]

Originally posted by: gsellis
I like REMHD. Recoil Enhanced Magneto HydroDynamics. Found out about that tech looking up MHD. MHD was being investigated in the 70's for power generation with biomass generation plants. MHD is designed to be significantly more efficient than just stream generation and use steam generation at the same time.

Paper on REMHD for space travel power

Big paper on Gen IV reactors Pg 91 - called Radiation Enhanced MHD here.

I know it's suppose to be a research paper, but some of the writting is a pain to read. One would think scientists are widely read, and be more elegant.

 

[spike spiegal]

On the other hand, the saftey mechanisms are implimented using so called "passive" saftey.

You mean, you object to the Russian Chernobyl design of a positive void coefficient?

Lets also make the tips of the control rods out of a weaker neutron absorbing material than the water the control rods are displacing.

 

[JohnCU]

i've worked in a nuclear plant for the past 3 summers, i can't wait for the new reactors to come online. less stuff that breaks, less systems to worry about. :thumbsup:

 

[BrownTown]

yeah, too bad the regulations are such a nightmare that it takes years just to fill out the paperwork, and then you get sued by all the enviros, and after you beat then they try and sabatoge you etc...

 

[Mark R]

Yup.

As an example - A plant operator in California wanted to upgrade one of their plants. There would be no alteration to the nuclear system. The only change would be to replace the exisiting steam turbine and generator with modern high-efficiency models.

The reactor would operate under the same conditions, the only difference would be a 1-2% increase in electrical output, and corresponding reduction in heat dumped from the cooling towers.

Permission was refused - the decision was that this upgrade was 'essentially the construction of a new nuclear power plant, and that this was illegal under state law'.

 

[Megamixman]

Pcy, the problem of nuclear waste has actually already been addressed. Yucca mountain is finally in its finishing stages. The only problem, is that all the nuclear waste we have built up over the years is enough to fill up Yucca Mountain.

This has lead to a huge change in US nuclear policy. Up till now, reprocessing of any nuclear waste has not been allowed due to the arms proliferation treaties, but amazingly enough Bush did something right and basically threw that out the door. Most of Europe reprocesses their Nuclear Waste and thus they have a very small amount of unusable nuclear waste. By reprocessing our waste we can extract most of the unused nuclear fuel, since in reality only a very small amount gets spent in the reactor and the rest ends up in the nuclear waste.

This coupled with the development of Phase IV reactors really should bring about a nuclear Renaissance for the US. Everyone blabs about Hydrogen fuel. but no one has a good way to make it yet. That's where phase IV reactors come into play. All power plants have some loss of energy due to the natural inefficiency of the conversion of energy. This is generally apparent in the coolant of the power plant. In a Nuclear Reactor the coolant is at several thousand degrees, unlike a normal combustion based power plant. By coupling the coolant to a facility located near the power plant that energy can be used to allow for a chemical reaction that produced hydrogen from water, but is only possible at very high temperatures. This increases the efficiency of the Nuclear Power plant.


Again the main problem is the bureaucracy. The Nuclear Regulatory Agency is highly understaffed, and i believe there are about 27 new reactor bids up. California, with its very high energy consumption really needs to make the first move and start to build more nuclear power plants. The problem there is the high level of scientific ignorance amongst many of the environmentalists. They are hard line advocates that drive atmosphere killing cars, and then complain about how the nuclear waste, which with the new policy should be cut by a huge percent.


About safety, don't forget that the laws of physics have been proven over countless years and repeated efforts. Active cooling, although powerful also brings in more point of failure. A system that can rely on passive cooling as a final safety is what will prevent major accidents. Don't forget there are probably active safety features to set the system down to a low enough state such that passive safety can effectively take place. It isn't like there isn't any active safety, it is just no longer the final safety point.



By the way, if you guys have questions, comments, rebuttals, etc. on what i said please feel free to Pm me. I'm only starting as a freshmen at Cal in the NE department, but ive been very interested in this area for about 2 years now.


PS. Do we have any NE's on here? Or NE students, besides me?

 

[BrownTown]

Well, most of the new reactors are likely gonna be built in the South which is funny since we already have the lowest energy prices to begin with. Of course California which already has some of the most expensive electricity will only get more expensive as they drive all the pwoer producers out of their state and rely on imported power or expensive natural gas. Power produced by natural gas or oil is many times more expensive than coal or nuclear, and using it as a base load source of energy is just ridiculus.

Its probably best to not even mention California in any "Highly Technical" discussion on energy policy since all they understand is enviromnetal alarmism and nonsensical regulation practices.