Hacker News Reader

Comment by DennisP | Hacker News Reader

A few years ago I got to sit in a meeting between reps from a bunch of GenIV reactor startups, and a former chair of the NRC (not Jaczko). The reactor people said their main problem was that the NRC required near-complete blueprints before they would even look at a design. Getting to that point took several hundred million dollars. Then the NRC would give a flat yes or no. If no then you were out of business.

That's a pretty difficult environment for investors. The reactor people said just breaking the process into phases would be a huge help. The NRC person was unsympathetic, said it wasn't the NRC's job to help develop nuclear technology, and was uninterested in climate change.

View on HN

So if I want to build a house I have to come to the building authority with blueprints and ask if that building design is OK. I can't just ask step by step: is it OK if I use these walls, these windows... And build up my house like that.

Why would that be different for a nuclear regulatory body? Moreover, they are designing new reactor technology much more complex and less understood than a house. I imagine for safety assessments one would need to look at the whole design, just look at the example on cable trunks given in the article.

I wonder what the startups would say if the agency would approve their first steps, but then disapprove the full blueprints, because even though the steps made sense the whole does not meet requirements. Or do they want guarantees?

I think all they want is to use modern, iterative development principles. It's fine to have the final 'yes' take the full lump sum and a 'no' being a real possibility. What they want is to be able to get multiple opportunities to fail and get a 'no' earlier in the process, when the stakes are smaller. While your building analogy is nice, I'll propose a different one - pharmaceutical companies face a similar, lengthy, expensive regulatory process, but a new treatment doesn't just go from a single prototype to human trials - there are many steps, costing more and more as you get closer to certification, and each can block the entire thing. We need a similar path for new nuclear designs if we want to be able to live on this planet for long.

It’s a political problem — congress is more interested in clean coal or fermenting grass.

Federal civil servants are trained early that their opinions are irrelevant.

Requiring blueprints would seem critical to actually being able to have a product to review for safety. Not only blueprints, but detailed material specs as well.

Nobody was asking for complete approval without blueprints. They just wanted an early indication along the lines of "sure, you're on the right track" or "don't waste your time, no way in hell we're approving anything like that." Or maybe "that sorta works but we would require X and Y," and then they could incorporate that into the design.

Reply on news.ycombinator.com