Hacker News Reader

Comment by helsinkiandrew | Hacker News Reader

One of the reasons for this is that every new nuclear power plant is a little bit different from all the previous ones - 'better', safer, more efficient etc., but different with more risk and expense.

A Solar plant is (very crudely) just purchasing panels from a factory that's already produced millions of them and installing them in a frame so they're pointing at the sun.

View on HN

I suppose, as a nuclear fan, if someone messes up solar panels really badly, the societal impact is low. If you mess up a nuclear power plant, it's pretty bad.

Not really, but each new reactor is treated like a completely new design which makes regulations incredibly expensive. It is actually very stupid.

Stupid the regulations or the design process?

Treating every reactor like it is a brand new design is very stupid, clearly.

Small, modular reactors [1] have been the dream of nuclear energy for a long time. The technology was proven a long time ago with nuclear submarines which have an outstanding safety track record. What’s stopping them from entering full scale production is a horrible mix of regulatory red tape and political opposition. Chief among the issues is a regulatory framework which is not designed to certify mass-produced reactors.

No one wants to invest the capital to get a manufacturing plant up and running if regulators aren’t willing to allow completed reactors to be commissioned and begin producing energy at a rate that matches the manufacturing rate. The insistence on treating every nuclear site as a unique project subject to years of environmental surveys and extensive, bespoke planning makes modular reactor designs moot as a technology. This is why every reactor gets a new design: they have to go through the process anyway so they might as well try to max out the production they can achieve at that site.

[1] https://en.wikipedia.org/wiki/Small_modular_reactor

Unlike SMR's the efficiency/economics of solar are not as dependent on scale. Even small solar installs can be productive.

What I mean by scale is scale of production (units per year). Solar wasn’t even close to economical when the first nuclear reactors were coming online. Solar became viable over many decades of production, technological improvement, and supply chain development.

Solar panel costs have fallen 3 orders of magnitude since the 1970s [1]. SMRs wouldn’t need that dramatic of a production cost improvement to become viable for commercial energy production. The main barrier remains regulatory.

[1] https://ourworldindata.org/data-insights/solar-panel-prices-...

I have yet to see good argument as to why the reactors on nuclear subs are "proof" of the viability of domestic commercial electrical reactors. They have different commercial viability, enriched fuel and safety requirements. Plus I don't think the primary use of reactors on subs is to generate electricity.

Not dismissing SMR for commercial use, but don't see how military subs / ships are relevant

> Not really, but each new reactor is treated like a completely new design which makes regulations incredibly expensive.

There are already-approved designs that are pre-approved can can be constructed without review:

* https://www.nrc.gov/reactors/new-reactors/large-lwr/design-c...

AIUI, one can get approval from the NRC (under Part 52?) by building a plant as-drawn with a pre-approved design:

* https://en.wikipedia.org/wiki/Combined_Construction_and_Oper...

* https://www.nrc.gov/reading-rm/doc-collections/cfr/part052/f...

If you deviate from the plan then you have to get approval for that deviation. Apparently this was part of the problem with Vogtle (Unit 3): they had a design and got it approved, but because the hands-on knowledge of building had atrophied, they found complications in construction in the drawings, and wanted to make change to simplify things. These changes then had to be approved, which add to delays.

There is another regulatory process where you get approval in phases, and this would have been better for Vogtle. The Decouple podcast has a four-part series on Vogtle that goes into a lot of detail about Vogtle:

* https://www.youtube.com/playlist?list=PLyouH0mkPJXHR0hKW_iLk...

One can get an initial site license without actually picking a particular design, so the process can start with decisions are being made:

* https://wyoleg.gov/InterimCommittee/2023/09-2023071715-03Nuc...

Isn't this the promise of SMRs?

They unfortunately haven't held that promise up.

Reply on news.ycombinator.com