Comment by njarboe | Hacker News Reader

Comment by njarboe | Hacker News Reader

Parent Full thread

The problem is that it is insanely costly and cumbersome to get regularly approval to build an experimental reactor in the US (and elsewhere I presume) so nuclear tech is stuck in the 1970's, just 20 years after the first commercial reactor was built. Imagine still using cars, or trains, or computers after only 20 years of development.

One has to design the whole reactor on paper before building it and get it approved by the Nuclear Regulatory Commission (NRC). If, when you build it, you find you need to make changes to the design that are above a certain threshold, you have to recertify (not sure of the details. Can't find a good link about this). Imagine trying to build something as complicated as a nuclear reactor and you can't make iterative improvements. NuScales design approval process, the approval to be able to build the first reactor, cost $500 million dollars, took 2 million man hours, included over 2 million pages of documents, and after submitting in Jan of 2017 did not get approval until around 4 years later in August of 2020[1]. This is just to be able to build the first design. No wonder nuclear power has seen no progress in the last 50 years.

Fortunately this company did persevere and now is planning to build the first power plant in Utah, hoping to be operational in 2030. The people/governments stating that climate change is a crisis, and I do believe there is way too much CO2 in the atmosphere, should be fast tracking this approved tech with as much money as usable to build thousands of these reactors as quickly as possible.

[1]https://www.nuscalepower.com/technology/licensing

There should be pretty strict regulation for fission reactor development, simply because radioactive materials in runaway reactions can have totally uncontrollable, millennia-spanning consequences. Very few other technologies have this potential - even a regular chemical explosion poses little risk to future generations in a large radius. Certain materials (plutonium, for instance) can reach criticality very quickly and in tiny quantities, and plutonium is a byproduct of fission. (I’m not a nuclear physicist, so my understanding of the real potential here might be wrong, because I have no sense of the quantities that might be experienced).

This isn’t a defense of the current regulatory process. The time frames, in particular, sound pretty egregious (4 years to approve a prototype, and 2m pages of documentation feels like a lot, too, but again I don’t know how these things are designed). But I would be extremely uncomfortable with lax regulatory oversight, given that nuclear accidents have permanent, irreversible impacts on society and geography.

Now, small-scale reactors may be a different beast entirely. If the quantity of materials is pretty much guaranteed not to have potential to cause problems for anyone but the operators for a short period of time…. Then there certainly seems to be a case for a shorter regulatory cycle. But I would be shocked if scale isn’t already taken into account for the current regulatory burden.

The US should set up a special zone for companies that wish to work with nuclear material and provide well designed containment labs to work in. I would nominate the nuclear test site in Nevada. That area was already used to test devices that purposely went super-critical and spread their nuclear material into the environment with the most powerful explosions man has ever created.

Those numbers are completely meaningless without comparison to how long it takes to get approval for other power plants. I could not find definite answers but this report [1] for Australia says it can take up to 15 years between original prospecting and operation of a windfarm. I found another source that said it takes on average 3.1 years for approval in Sweden (it was not clear if that applied to windfarm only). Several other sources talk about multi year time frames as well. Considering the comparable impact of nuclear vs e.g. wind 4 years is quite short. Also this also disproves that nuclear is 3xpensive because of regulation, other energy sources face similar regulation delays.

[1] https://www.nwfc.gov.au/observations-and-recommendations/cha...

What I am talking about is the approval to build the first version of the reactor. The demo reactor. Imagine if one had to wait four years and spend half a billion dollars to get approval of your design of a windmill (not windmill farm) before you could even build the first one to see how well it would work. Progress on windmill design would be slow.

Actually building the power plant with the reactor at a specific location is a different problem.

Why can't these designers set up their experiments in a country that has lax or no regulation? I can understand security concerns but surely there is some part of the world amenable to these kinds of experiments

Reply on news.ycombinator.com