Originally Posted by JasonC SBB
Re: nuclear costs - how much of it is due to badly written gov't regulations?
In the US? Probably > 75%. I'm not sure I'd condemn the whole NRC as "bad", but it's definitely a hell of a bureaucracy.
Back when nuclear power was first being talked about, it was supposed to be "Too cheap to meter" (Lewis L. Strauss, chairman U.S. Atomic Energy Commission, September 16, 1954)
Trouble is, I think ole' Louey was only looking at the technical aspect of the problem, and not the political one. What if nuclear reactors in the 250-500MWe range were built in a mass-production environment and type certified rather than site certified? My guess is that while electrical power might still merit metering, the end-user cost would be a fraction of what it is today. And that doesn't even account for the secondary and tertiary costs of not
having nuclear power; environmental impact from fossil fuels (and the money spent on it), dependence on foreign energy supplies (and the costs, both direct and military of that), R&D spent on blue-sky renewables that will never be economically viable, etc.
Originally Posted by JasonC SBB
Greenies turned it around and said "nuclear, bad!".
Well, it didn't help that the film "The China Syndrome" had opened in theaters just 12 days prior
, and that in it, Jane Fonda noted that an accident of the exact sort which did happen at TMI would "render an area the size of Pennsylvania permanently uninhabitable."
(TMI, for those who don't know, is in fact located in Pennsylvania.)
So I don't really blame the greenies, at least not for the immediate hysteria. I blame Hollywood.
Chernobyl didn't help either. That one genuinely was a catastrophe, and again, mostly the fault of poor reactor design.
Originally Posted by JasonC SBB
3 Mile Island was a demo that the engineers designed it right. It failed safe despite the crew's best efforts.
Well, yes and no.
Good engineering is what prevented a major accident from turning into an outright disaster after it had happened. The reactor vessel remained intact and containment was not breeched.
OTOH, bad engineering is what allowed the meltdown to occur in the first place. Here's a very brief summary of what happened:
1: Some dipshit maintenance worker cross-connected a high pressure water line to the plant's process air system, presumably by accident. (Bad engineering: both the service water system and the safety-critical process air system used the same physical connectors.)
2: Said water started working its way down the air lines until it hit the flow control valves on the secondary cooling system (secondary meaning the loop that transfers heat from the boiler to the turbine, not implying that it was secondary in importance.) When this occurred, the valves failed shut. (Bad engineering: the valves should have failed in their last-known state, not shut.)
3: When the valves slammed shut, the turbine automatically tripped (good) and the reactor scrammed (good) however the resultant water-hammer ruptured the **** out of the secondary loop. Think about compressor surge when you slam the throttle shut with no BOV, only with tens of thousands of gallons of very hot water and steam. For reference, the secondary loop is the principal method for removing heat from the reactor. So at this point, there is no longer heat being drawn out of the reactor, and it is no officially becoming a very bad day, though the situation is not unsalvegable.
4: As a result of #3, temperature and pressure in the primary loop starts to skyrocket. (For reference, the primary loop is the one that flows through the core itself, carrying heat to the boilers where it is exchanged with the secondary loop. The primary loop is radioactive.) A relief valve, called the PORV, on top of the pressurizer (the highest point in the primary loop) opens to vent steam into a collection tank, which is inside the containment building, in order to prevent the primary loop from rupturing. (good) Unfortunately, this particular valve was known to be a shitty design that tended to stick open from time to time. It did, and continued to vent and vent and vent, overflowing out of the tank and onto the floor of the containment building. (Bad engineering: if you know that a particular part is a piece of ****, replace it with one that isn't.)
5: There is, inside the control room, a big-*** light which indicates the status of this valve. At least, that's what the operators thought
it indicated. It actually indicated the commanded status
of the valve, meaning what the valve had been told to do, not what it was actually doing. So, the big-*** light goes out, the operators think the valve is closed even though it isn't. (Bad engineering: if a $5 part is capable of causing a meltdown, at least instrument the damn thing properly.)
6: Because of this valve, steam continued to vent inside containment, lowering the pressure in the primary loop below the boiling point, which caused the water inside the core to start to boil off. (This is the beginning of the meltdown proper.)
7: The Emergency Core Cooling System automatically switched on, to directly flood the core with cool water. Unfortunately, all
of the service valves on the ECCS were closed, presumably as a result of some maintenance work that had been done in the weeks prior. This was a violation of operating procedure, but the automatic control system was not designed to detect this unsafe condition. (Bad engineering: the reactor should not have been permitted to operate with the ECCS inoperative. The valves were monitored, but only by small indicator lights on the panel which the operators didn't notice. The reactor's control system should have been interlocked to these valves, causing an auto-shutdown any time all of them were closed.)
8: The operator's only indication of the water level in the primary system (including the water level in the reactor itself) was the water level in the pressurizer, which is basically a holding tank near the top of the primary loop. (The same as mentioned in #4.) Because water was boiling in the core, the water level in the pressurizer went UP, despite the fact that they were loosing tons of coolant through the PORV, none was coming in from the ECCS, and the water level in the core was dropping. So, the operators not only shut down the ECCS pumps, but also started draining
water directly out of the primary. (Bad engineering: there was no way for the operators to know that water level in the core was falling, only that the water level in the pressurizer was rising. This led them to take the wrong actions.)
9: The primary coolant pumps (which are massive bus-sized electric motors) start going overspeed and vibrating like hell as they begin picking up the steam that's being produced inside the core. The operators shut those down as well, to prevent them from rupturing. Their training tells them that convection will cool the reactor, however it can't because steam bubbles are blocking the pipes. And remember: they still think the core is full of water.
10: At this point, the reactor really starts to go apeshit. With all flow stopped, coolant is now explosively boiling in the core, and still pouring out of the PORV. This is when the fuel rods become uncovered, and subsequently melt. As they do, the zirconia cladding of the rods reacts with the steam to form hydrogen gas, which subsequently boils out through the open valve, into containment. (Bad engineering: try to make your fuel rods out of something that doesn't create an explosive gas when it breaks down.)
TMI2 is now officially down for the count.
11: Finally, one of the operators, going through the board methodically, discovers that the ECCS valves are shut. He opens them and re-starts the pumps. It's too late to save the plant, but fortunately, the molten fuel stopped at the bottom of the reactor vessel and didn't escape, so no China Syndrome. (Good engineering? Who knows what would have happened if they hadn't finally restarted ECCS.)
The core begins to re-fill (although they have no way of knowing it) but they still have zero pressure and uncontrolled boiling.
12: Shift change. Fresh operators arrive in the control room.
13: One of the new operators, looking at the situation with a clear head, realizes that nothing makes sense, and at least some of the instruments have to be lying to them. Noting that both temperature and pressure inside containment are higher than they ought to be even under the circumstances, he checks a temperature sensor in the blowoff line after the PORV, finds it off-scale high, and surmises that the PORV has failed open. He closes an aux valve, which stops the venting. Pressure begins to stabilize, collapsing all of the steam bubbles that have been forming.
14: All of the hydrogen gas that got formed in #10, and has just been hanging around inside the containment building, explodes. The containment building does not rupture (good engineering)
, however the operators decide that in order to bring the pressure in containment under control and prevent a rupture, to vent some of the pressure from inside it directly to the atmosphere. This is the only release of radiation that occurs.
15: After much jury-rigging, the primary system is finally purged of all steam and the primary loop pumps are re-started, dumping their heat into a secondary secondary system. (No, that's not a double-negative.) The crisis is now over.
So yeah, good containment engineering saved the day. Shitty process control engineering caused the whole mess in the first place.