Hacker News Reader

Comment by version_five | Hacker News Reader

I don't think it's stagnant. I continue to see stuff like small modular reactors making progress. The obvious reason why it doesn't get more play is because it doesn't serve anybody's political goals. Renewables and carbon taxes or other measures to try and reduce quality of life (or opposition to those) get attention because they have good political value. Actually spending money and generating sustainable power, not much you can do with that, it doesn't match an ideology. Same with carbon capture and storage. There are good technical motivations to climate change, they just don't have a political champion.

View on HN

Small modular reactors don’t actually solve any significant issues for the nuclear industry. Construction costs are dwarfed by operating costs and the idea that N small reactors with N times as many parts will somehow require less maintenance is wishful thinking.

They very well could be slightly cheaper over the first 15 years, but nuclear reactors are designed for ~50 year lifespans and are really expensive to decommission.

Look at say Palo Verde it’s close to 4GW with a 82.80% lifetime capacity factor, but needs 2,055 full-time employees and that’s the best case.

That's totally wrong about operating vs construction costs.

Nuclear plants are cheap to run once you have them built, but they are expensive to build when everything goes right and frequently they don't.

There has been a lot of hand-wringing about the costs of the nuclear fuel cycle and frankly it's a bit ridiculous. For instance, they were expecting Yucca Mountain to cost $100 billion to run back when they were planning to run it.

People look at that and think "that's a lot of money" but a nuclear reactor makes about $500 million worth of electricity a year so that is 2 years worth of electricity from the 100 reactors that operate in the US.

(It's more absurd that we'd bury nuclear waste in Yucca Mountain when 98% of the energy content of the fuel remains in the waste! A reprocessing cycle would cost "more" but not much compared to the capital cost of building the reactors.)

> cheap to run

2000 staff represents well over 10 billion dollars in just salary over 50 years. Add in replaced equipment, fuel, insurance, etc and it’s a long way from cheap.

The assumption they are cheap to operate simply doesn’t hold up even the most basic investigation.

4 GW * 50 years is about $100 billion worth of electricity.

That's not the point, construction costs should be ~< 8 billion. current mini reactors cost 1 billion per unit (although I expect those costs to shrink) However 2000 employee is absurd and only a result of insane paranoid regulation and also because those plants were designed before modern computer science, therefore they probably underexploit sensors, and semi-autopilot abilities or autotuning. Russian models are actively reducing the needed number of employee for maintenance: construction cost: Specifications include a $1,200 per kW overnight construction cost 54 month planned

requiring about 35% fewer operational personnel than the VVER-1000

This was for VVER 1200 however the new VVER TOI will be the first to use software? > United information space is a multi-platform software and hardware complex intended for control of the engineering data for engineering, designing, as well as arranging communication between territorially-distanced project participants.

edit: can someone help me calculate the construction costs of a VVER 1200 ? it output ~1200MWe of energy Specifications include a $1,200 per kW overnight construction cost if 1 MWe is 1000Kwh (don't know if we're talking per hour..) then total construction cost would be 1200 * 1000 == 1200000 dollars 1200000 is a mere million, that doesn't make any sense, how to calculate? https://en.wikipedia.org/wiki/VVER#:~:text=Specifications%20... edit I was dumb the equation is 1200 * 1000 * 1200 == 1440000000. Therefore, if that is to be believed, a VVER 1200 produce 1200MWe and only cost 1.4 billion in construction costs (without the small interest rate and the operating cost) That is extremely cheap, the pathetic and hyped small reactors from rolls royce cost 1 Billion per unit and produce < 200 MWe... That makes the VVER 1200 6 times cheaper and probably the cheapest energy source on earth (again assuming those numbers are true), and the VVER TOI successor should be even cheaper given it is constructed in 17 months less.

No western country will be touching Russian nuclear power plants anytime soon. Maybe they could go for a Korean plant though. And yes, SMRs likely have an even bigger economics problem than regular sized plants but at least the upfront capital costs are lower and that makes (private and state) investors happy.

> the idea that N small reactors with N times as many parts will somehow require less maintenance is wishful thinking.

Not sure that's necessarily true. Building a lot of something allows you to learn fast and update designs for higher reliability/maintainability. Building smaller units could also free up costs that go into current designs, just because of their size. Certainly designing for zero-staff, eliminates one big source of unreliability.

Most SMR are sealed designs that can't be opened before their 20 year lifetime is over.

Then you need three of them or to refurbish the first one twice to reach the equivalent of a 50 year lifespan. At which point construction costs again smaller than operating costs.

It’s the same thing with steam turbines, they last a long time but not 50 years. So initially they are part of construction costs but eventually they are part of maintenance costs.

Essentially 50 years is used because cooling towers, buildings, plumbing, etc can last at least that long so decommissioning early is a waste. But eventually repair or replacement gets too expensive.

Carbon taxes are something separate from the technology. In fact, carbon taxes would drive adoption of better technology.

For instance it is completely practical to run the output of a fossil fuel power plant through

https://en.wikipedia.org/wiki/Amine_gas_treating

compress the CO2 to 1500 psi and inject it underground into a saline aquifer. Hardly anybody does it because it's expensive and nobody pays them to do it.

If there was a carbon tax that made it uneconomical to not do that, or if there was a subsidy for pumping carbon underground then people would do it.

It is all the same for nuclear power, extensions of renewables and other technologies that aren't profitable on their own.

It is a deal with the devil however to do so because it is a withdrawl from the government's legitimacy bank account. It's certain that any carbon trading system is going to lead to a few carbon traders getting rich and them feeding back 1% of their profits to politicians to keep their privileges. It's less certain that the planet gets saved.

It seems to be a more realistic plan to develop a technological revolution that really is cheaper than the alternatives... Because then you've saved the planet and the job is done

https://www.moltexenergy.com/

Small modular reactors are not even small. They're 400MW each and size-wise they conveniently don't include the containment building and all the supporting equipment and machinery. I don't know who came up with that name but it's not very descriptive term.

A potential SMR power plant from NuScale contains 5 SMR that each can be individually replaced. You forgot to qualify your "each". Each power plant can do 400MW but each individual SMR only does 77MW.

Compare that to conventional nuclear reactors where a single reactor does 400MW and the entire plant does 1.6 GW

Nuclear power is a terribly difficult engineering problem. You can't just put a nuclear power plant in your backyard. You can do that with wind or solar though.

Carbon taxes are hated by average citizens more than nuclear power because of the evil tax word.

Reply on news.ycombinator.com