Sunday, September 26, 2004

Is Nuclear Power a Good Thing?

Let me start out by saying I'm intimately familiar with nuclear power on a technical basis. I was trained in the Naval Nuclear Power Program. Probably the best nuclear power training in the world. Like any military training a lot of effort went into dealing with "accidents" other wise known as enemy action.

As a commercial power generation technology it is not viable.


Because it is uninsurable. Now why is that? The risk (at least in America) is low. Why can't it be insured? Because low risk is not zero. And one of the non-zero risks is a wide spread release of radioactive material. A risk in the trillions worst case. Now you can insure a coal plant because even in the event of its total destruction you are only talking billions in damage. Given the size of the insurance pools and the low risk an acceptable bet.

Nuclear power is only viable with government insurance. If the loss exceeds the risk pool guess who gets to pay? All us taxpayers. So that is reason #1. The risk/reward ratio is not commercially viable.

Reason #2. Commercial plants in order to keep the $$$ flowing might be willing to take short cuts. Read some of the NRC incident reports. They make my hair stand on end. Why? They only evidence the usual human stupidity. Which is just the problem. Stupidity is usual. Even for people as bright and well trained as reactor operators. So money stands in the way of safety (which is why Naval Nukes are less of a problem - the military has money to spend on safety that would not make commercial sense).

Reason #3. They do not make economic sense. Wind turbines are coming down in price. When turbine size (on land) gets in the 4 - 5 MW range, wind will cost the same at the bus bar as nuke or coal fired plants. A 5 MW (peak) turbine well sited is good for about 1 MW of base load (if you have enough of them to statistically average). So even though wind is intermittent if you have enough turbines you get a significant base load capability. Generally over the period of a year a well sited turbine will supply about 35% of its rated capacity on a continuous basis. One the turbines get larger than about 5 MW the electricity from them will cost less than coal or nuke electricity. By the time wind becomes a significant part of the grid (above 20%) we will have commercially viable electrical storage. Super flywheels.

Reason #4. Nuclear proliferation issues. All nuke plants produce plutonium. Power plants are refueled around every 18 month to two years. The fuel removed is currently being stored at the plants in water filled pools. If we ever start reprocessing those spent fuel rods we are going to have a lot more plutonium on our hands. A whole lot more. That means we have to inventory and guard it for the next 10,000 years or more. We haven't been doing technology long enough to be assured that is possible. The risk is difficult enough for military purposes. Do we really want another 100 or 500 reactors pumping out plutonium?

And yes I know about burning the stuff up in special reactors - which are devilishly hard to control for reasons too technical to go into here. I will give a pointer for those interested: prompt vs. delayed neutrons. In any case let me just say that the safety issues in a plutonium burning reactor have not been solved to my satisfaction, considering that normal nukes are uninsurable.

Nuclear power is not coming back to America. No matter how progressive your stance on the subject. The two main reasons - it is too risky. It is uneconomical.

BTW in my day job I help design and test aircraft AC and DC power generation systems. The criteria for the design is one unrecoverable failure every billion hours of operation. Given that a fleet of a given design has 1,000 aircraft and the aircraft operate (equivalently) 200 days a year, 24 hours a day, that works out to one total failure every 200 years or so. An acceptable risk. When you think of all the parts that go into such a system a truly amazing level of performance. Now think of a nuke plant with the same level of risk for total failure. The risk is unacceptable.


David Foster said...

What do you see as the driving force for cost reduction for the super flywheels? Seems that motors, generators, and gearing are all already fairly mature technologies; what else is in there other than the flywheel rotor itself?

M. Simon said...


You are exactly correct and that is why I'm so optimistic.

There are three places that need work.

1. The flywheels themselves
2. Very low friction bearings (probably magnetic)
3. Production tool up

My main question is: as promising as this technology is why aren't the telcos dumping cash into this to get rid of their batteries? There is some kind of disconnect and I'm not sure of what it is. I'm going to research it and write something up.