Hacker News Reader

Comment by mountainriver | Hacker News Reader

As a complete novice, this could be a game changer?

The ability to centralize the production I imagine radically reduces the cost, or at least has the potential to.

With all the energy challenges we face, could the US government subsidize a program like this and make it a silver bullet?

View on HN

Yes. When I attended an IAEA Safeguards conference in 2018, there were a number of these civilian nuclear battery designs but none had been built. A small nuclear battery can replace building-size diesel generators which remote communities and islands, data centers, etc. currently use as a primary or backup power source.

Depends what it costs. Since the article neglected to give a figure, Im pretty skeptical it's going to be competitive.

This video [1] talks about some of the potential cost savings and reduction in time to deployment. There is one being installed near me so I guess time will tell how realistic the projections are.

[1] - https://www.youtube.com/watch?v=xxXlD4e-wTE

The most reliable output of the nuke industry has always been dishonesty. There is no reason to expect this will change suddenly, or that it has already.

$ 3 billion for 720 MW of power (2019 figure) [1]. More recent figure of $ 5.5 cents per KW/h [2]. So about 4x cost overrun leeway to be competitive in Europe. That seems manageable?

[1] https://www.science.org/content/article/smaller-safer-cheape...

[2] https://www.nucnet.org/news/first-customer-has-set-lcoe-targ...

No. It is not.

The US govt is massively subsidizing this whole venture, and none of that cost to taxpayers is figured in.

You’d need to compare to the subsidies and externalities of other power sources then right?

Solar panel imports are even slapped with tarriffs. It's unlikely if you compared like for like it would come anywhere near the subsidies nuclear demands.

Installation is subsidized and you need to massively overbuild without storage or otherwise build storage.

Overbuild of solar and wind has become cheap. But overbuild required is nowhere near that commonly assumed.

Storage will of course be built too, at overwhelmingly less cost than nukes, and much, much faster. But, first, the generation capacity to charge it up from.

You still have the problem of transmission, and grid scale storage is basically not deployed anywhere outside of a pilot program in Australia right now.

You not knowing about a thing is not the same as it not existing.

The cost of global warming destroying human civilization should be built into the actuarial costs here as well.

> costs

1 Fukushima is too many, no?

No. Fukushima is a price I'd gladly pay and I expect every reasonable person to agree.

The alternative isn't "no Fukushima", the alternative is hundreds of thousands deaths per year by burning coal. It's just not "one huge bang" so people don't realize it, because understanding abstract dangers is hard.

Just improve decentralized energy storage and micro-grids.

Absolutely.

The Fukushima disaster: https://en.m.wikipedia.org/wiki/Fukushima_Daiichi_nuclear_di...

The earthquake: https://en.m.wikipedia.org/wiki/2011_T%C5%8Dhoku_earthquake_...

Zero known fatal injuries out of 20,000 were caused by the reactor.

If I've done the math right, a nominally operative coal plant over Fukishima's 40 year lifetime would have caused 5ish deaths.

Fukishima did cause a few cases of cancer, but so does nominally operative coal.

People can gripe about tail risks from environmental pollution when it's not displacing a worse modal risk. Right now it's unreasonable to. Proliferation risk is concerning; most of the rest is just scale insensitivity.

Nukes' competition is not coal. It is renewables. Comparing to coal is tendentious and disingenuous, not to say dishonest.

Not really m. Something has to provide power when the sun and wind aren’t available. Right now that’s natural gas, coal, hydro, or nuclear.

"Right now". Right now all the nuke plants you are talking about don't exist, so produce exactly 0 watts.

We’ll there is a good deal in some places.

For now, until their production cost exceeds alternatives. Then they will be mothballed.

Pumped storage and grid scale batteries provide that at much cheaper cost overall.

Nowhere near as much is needed as people think either.

Where are grid scale batteries doing that today?

Before a thing has been built, it needs to be built. Do you need this explained further?

The overwhelming bulk of utility storage will not be batteries, because that is the most expensive alternative. Utility storage will be whatever is cheap and locally practical.

That’s basically what I meant.

California, Hawaii, Australia.

In hawaii they canceled a proposed inter-island power connector because batteries + solar were just cheaper.

Pumped storage is generally a lot more economic and can store a lot more than batteries but it cant be built quite as quickly & is somewhat geographically dependent. It takes 4-5 years to deploy rather than months.

that assumes grid scale batteries exist that can back up power for 10-16 hours a day at a time and those don't exist.

You should look up the cost (in whatever metric you like: dollars, lives, cancers, etc.) and compare it to coal. Don't forget to normalize, since coal is much more common than nuclear power.

The results will likely surprise you.

It cuts both ways. You lose the scale, but its more modular.

Putting a conventional PWR in such a modular system isn't a silver bullet and has you to be proven to actually be cheaper and a game changer.

If you simply want 1.5GW it might be simpler to just put a single PWR there rather then like 5 of them.

I would say real GenIV modular reactors are the silver bullet, this is a step in the right direction.

I learned on HN that we almost did this in the 1970s, albeit with large but modular reactors. The effort extended far into construction of the (very large!) factory facility.

https://news.ycombinator.com/item?id=31467070

Yes, this is huge, IMO.

Reply on news.ycombinator.com