[bendykstra]

There were other issues as well.

Decommissioning[1]

With a total of 438 malfunctions during its operational lifetime from 1964 to 1972, the reactor at McMurdo proved to be an unreliable source of power generation, available only 72% of the time. Because of this, in May of 1972, it was determined by an independent assessment from the Bechtel Corporation that the PM-3A was not cost effective and that replacement with modern diesel-electric generators would require less staff and would also be more reliable.

1. http://large.stanford.edu/courses/2014/ph241/reid2/

[jessriedel]

It's really a shame there's so little innovation going on for these sorts of devices. There's nothing inherent about a nuclear device that will make in unreliable.

[jfoutz]

Eh, I'd take a mass produced object over a hand made version any day. Even more important if reliability is an issue.

Ferraris are nice and all, but a boring old camery will get you where you need to go for hundreds of thousands of miles.

It makes a lot of sense to me to just use what hospitals and data centers and military bases use. Also, if one breaks, you can probably use it for parts for the rest.

[eru]

That ideas was actually proposed to revive nuclear reactors: instead of building them on site, come up with some energy-generating components to be mass produced in a central factory and shipped out.

(You can make those parts inert and safe to handle. Even the fuel rods of current reactors are safe to touch before they are spent.)

[jfoutz]

IIRC, in France every reactor was the same model. If an issue was discovered, all of them would be upgraded to handle the problem. It also allowed a little more workforce mobility. It is a good idea.

I just don't see how we can get there from here. Nuclear isn't particularly cheap (but yes, a good chunk of the expense comes from regulation). So there's no obvious reason to choose that over any alternative. We could certainly change tax structure to force oil and coal to be more expensive, but even simple changes to account for externalities are met with massive massive resistance.

[eru]

Pricing the exernalities of oil and coal into them might be quite a good idea---especially since it encourages boring things like just using less heating or driving less. Those boring options are seldom encouraged in `politician picks the winner in a beauty contest' subsidies.

But yes, we won't see that..

I wonder why noone has ever tried selling eg a carbon tax (or any kind of broad and useful tax) in combination with basic income? If you earmark the windfall from a carbon tax to be equally shared by all residents, you get quite a few natural allies and take the wind out of the sail of arguments along the lines of "think of all the poor people".

[mschuster91]

> Even the fuel rods of current reactors are safe to touch before they are spent

What? Fuel rods consist of uranium, which alpha-decays into thorium (a sheet of paper is enough to shield the alpha particles), but the next decay step is beta-decay to thorium, and that needs more shielding. No way one can touch a fuel rod without getting a load of radiation.

[Hondor]

The half life of U-235 is hundreds of millions of years, and U-238 even longer. So it's not decaying fast enough to be dangerous.

[jfoutz]

Yeah, maybe there's some other model, but there would be no reason for a Spent Fuel Pool [1] if you could handle the things.

There's also an interesting anecdote about spent fuel here [2]. You can get pretty close, if it's under water... but it gets very dangerous very fast.

[1] https://en.wikipedia.org/wiki/Spent_fuel_pool [2] https://what-if.xkcd.com/29/

[eru]

Spent fuel is dangerous to handle. That's why I specifically mentioned the unspent fuel.