News for hackers. Stuff that matters.
Re:NRC Certifies First U.S. Small Modular Reactor Design by PaulHoule on Friday January 20, 2023 @03:01PM (#34458959)
Have you seen how big the steam turbine is?

There is a table here that estimates that only 28% of the cost is the "nuclear island"

https://world-nuclear.org/information-library/economic-aspec...

the steam turbine and other systems that are bloated by low temperature overhead comprise much of the rest. Also some of the "nuclear island" such as the steam generators is also bloated by low temperature overhead.

It is no accident that we stopped building coal-burning power plants at the same time we stopped building LWRs and that is because gas turbine power plants with much lower capital cost became available.

Re:NRC Certifies First U.S. Small Modular Reactor Design by Retric on Friday January 20, 2023 @03:05PM (#34459026)
28% of construction costs not total costs. If we are assuming magic such that you don’t need fuel then you also don’t need armed security, nuclear decommissioning etc etc.
Re:NRC Certifies First U.S. Small Modular Reactor Design by PaulHoule on Friday January 20, 2023 @03:29PM (#34459397)
A story I've been gathering bits and pieces of evidence for but haven't put together completely is that nuclear decommissioning projects, unlike construction projects, frequently end up completed ahead of schedule and under budget.

This is even true in cases where the situation is unprecedented and people are having to develop new techniques.

Re:NRC Certifies First U.S. Small Modular Reactor Design by Retric on Friday January 20, 2023 @03:50PM (#34459745)
As is so often the case it isn’t any one issue like decommissioning that’s the problem alone. It’s that nuclear has such a wide range of costs that they collectively become expensive even if each cost in it’s own isn’t prohibitive.

Aka if we only needed to pay for fuel rods and waste management then nuclear would be wildly profitable. Similarly if the only cost was a large workforce and expensive maintenance then again it would be wildly profitable. Being forced to act as base load generation with long periods offline for refueling isn’t a deal killer. If it was just the long construction times and NIMBY issues that would be fine. Etc.

Unfortunately because there is such a diverse range of costs there isn’t a single silver bullet that’s going to solve all problems with nuclear power. At best by addressing individual issues we might increase the percentage of electricity generated by nuclear power. That’s very realistic and IMO a worthwhile goal.

Re:NRC Certifies First U.S. Small Modular Reactor Design by arcticbull on Friday January 20, 2023 @03:10PM (#34459104)
Construction costs are the big cost in nuclear power. Uranium comprises between $0.0015/kWh and $0.000015/kWh.

You don't really need armed security, but a couple guys with guns in America are a dime a dozen.

Decomissioning is $300-400M after a 30-50Y lifecycle and operators are generally allowed to collect that money over the plant life. [1] That's compared to the $17B in construction costs for Vogtle.

[1] https://www.nrc.gov/reading-rm/doc-collections/fact-sheets/d...

Re:NRC Certifies First U.S. Small Modular Reactor Design by Retric on Friday January 20, 2023 @03:16PM (#34459205)
Uranium costs are the kind of meaningless fact that’s true and wildly misleading. Fuel rods are not simply long sticks of unprocessed uranium.

Refueling is expensive because of many separate costs. Even simply being forced to take a power plant offline for a long period is inherently expensive. Similarly building a cooling pond and equipment to move extremely high level nuclear waste is costly. Add up all those individual costs and fuel represents a significant faction of the total lifetime costs for a nuclear reactor.

Re:NRC Certifies First U.S. Small Modular Reactor Design by moloch-hai on Saturday January 21, 2023 @10:17AM (#34466963)
We don't actually know what most decommissionings will cost. $300M is just a lower bound.

Nukes' fuel cost seems low only in comparison to their other very high costs.

Re:NRC Certifies First U.S. Small Modular Reactor Design by arcticbull on Saturday January 21, 2023 @03:58PM (#34470449)
> We don't actually know what most decommissionings will cost. $300M is just a lower bound.

Of course we do, we've decommissioned plants before. About 200 commercial and 500 research reactors. That's a sufficient sample size. [1]

> Nukes' fuel cost seems low only in comparison to their other very high costs.

$0.0015/kWh is objectively cheap on an absolute scale. I don't know if you noticed, but California pays about $0.19/kWh, so this would be 0.78% of the delivered cost.

[1] https://world-nuclear.org/information-library/nuclear-fuel-c...

Re:NRC Certifies First U.S. Small Modular Reactor Design by Retric on Saturday January 21, 2023 @04:18PM (#34470650)
$0.0015/kWh is actually quite expensive.

That’s ~7% of the total cost for solar per kWh and doesn’t even get you to fuel rods.

Re:NRC Certifies First U.S. Small Modular Reactor Design by arcticbull on Saturday January 21, 2023 @04:20PM (#34470659)
The fact it's expensive on a relative scale is irrelevant because it's cheap on an absolute scale. In fact the price of raw uranium input is 100X lower than that in a breeder reactor because more of the fuel is consumed, $0.000015/kWh. Being objectively cheap also allows you the flexibility to collect it in more expensive ways, for instance seawater extraction (which makes nuclear renewable) is only double the price, and falling as the technique is improved.

> That’s ~7% of the total cost for solar per kWh and doesn’t even get you to fuel rods.

Which again doesn't matter because it's still objectively and on an absolute scale very cheap. About 0.8% of the cost you pay for electricity.

Solar is great, nobody, certainly not me, is trying to tell you not to build solar.

The reality is the future of the grid is going to be a mix of generation sources. That's going to include solar. It should, in my opinion, also include nuclear due to their different generation characteristics.

> The only thing that matters in economics is the relative scale. Nuclear being more expensive compared to the alternatives is a deal killer even if it’s not that expensive per kWh in absolute terms.

I disagree because it has different supply characteristics. One supplies a constant amount over a long period of time and is difficult to adjust. The other varies massively over the course of a day and zeroes out at night. Solar alone isn't going to meet needs, you need either or both of base load plus storage in addition.

It's disingenuous to compare the price of a kWh of solar by itself to nuclear when one works at night. If you want power at night, which I think many of us do, then you need to price into the $/kWh rate the cost of storage. You need to compare like for like.

[edit] In 2021, utility scale solar-plus-storage with a capacity of 50 MW/200 MWh is estimated to reach $0.085-$0.158/kWh. Nuclear is $0.131-$0.204/kWh. [1] They're actually quite comparable, and we have line of sight to making nuclear cheaper. Again, nobody is advocating for a 100% nuclear grid, it's not possible, because it's only suited to providing base load. A 100% solar grid is impossible because of the night time.

[edit2] Are you not reading what I'm saying? Nobody is advocating for a 100% nuclear grid. I am advocating for a mixed grid of renewables and nuclear where each operates according to its optimal utility function.

[1] https://www.pv-magazine.com/2021/11/05/utility-scale-solar-r...

Re:NRC Certifies First U.S. Small Modular Reactor Design by Retric on Saturday January 21, 2023 @04:24PM (#34470699)
The only thing that matters in economics is the relative scale. Nuclear being more expensive compared to the alternatives is a deal killer even if it’s not that expensive per kWh in absolute terms.

Again if 7% represented a total fuel costs that might be helpful benchmark, but Nuclear’s actual fuel costs are higher than Solar’s total costs.

Yes, that’s right even if the only costs where fuel rods nuclear would already be more expensive than Solar per kWh.

Edit: In response to your edit unsubsidized Nuclear is currently more expensive than unsubsidized solar + batteries which can not only provide 24/7/365 power but actually respond to changing grid demand. Base load power isn’t a benefit it’s a major limitation to adoption because demand isn’t constant.

A true apples to apples comparison shows a 100% Nuclear grid would required vast price increase, while a 100% solar grid is roughly the same price as what we pay today.

Edit2: “Nuclear is $0.131-$0.204/kWh” that’s only for base load nuclear costs skyrocket if you want to respond to changing grid demand on a 100% nuclear grid. Rough estimates are close to 50c/kWh for a pure nuclear grid which is why nobody did so and even France was forced to import and export a large fraction of their generation and useage.

Re:NRC Certifies First U.S. Small Modular Reactor Design by moloch-hai on Saturday January 21, 2023 @04:43PM (#34470896)
There is no place for nukes, just because because they cost so many times more than renewables. They cost, per kWh, even more run intermittently.

What you need in backup generation is cheap construction. Storage cost is falling even faster than solar and wind.

We already have the combined-cycle gas turbines. Their opex falls with duty cycle. Duty cycle falls with renewable generation buildout and, eventually, storage buildout.

Re:NRC Certifies First U.S. Small Modular Reactor Design by Retric on Saturday January 21, 2023 @04:56PM (#34471008)
> [edit2] Are you not reading what I'm saying? Nobody is advocating for a 100% nuclear grid. I am advocating for a mixed grid of renewables and nuclear where each operates according to its optimal utility function.

The point was illustrative. We already have nuclear, the only point of advocating nuclear is if you want to increase it. Unfortunately, a 35% nuclear grid costs more per kWh than a 30% nuclear grid, and I don’t think most nuclear advocates understand why.

Your 50 MW/200 MWh numbers are basically what it costs for a 100% solar grid. So it’s not even clear if any nuclear would be cost effective in most areas. Alaska and Russia clearly can benefit from nuclear power, it’s not obvious if California, Texas etc will.

Re:NRC Certifies First U.S. Small Modular Reactor Design by pfdietz on Sunday January 22, 2023 @03:29AM (#34475470)
It's not obvious Alaska can benefit from nuclear. The largest grid in Alaska, the Railbelt Network, has an average power flow of just 600 MW.
Re:NRC Certifies First U.S. Small Modular Reactor Design by Retric on Sunday January 22, 2023 @06:55AM (#34476494)
The article is talking about small modular reactors, assuming the price per kWh is similar for 3x50MW as it is for current 1GW designs they should be able to lower the 18c/kWh average prices. https://www.electricitylocal.com/states/alaska/
Re:NRC Certifies First U.S. Small Modular Reactor Design by moloch-hai on Sunday January 22, 2023 @12:34PM (#34479876)
Alaska has wind. And also very long summer days.

Solar has got so cheap that idling them half the year is no big deal.

Re:NRC Certifies First U.S. Small Modular Reactor Design by arcticbull on Monday January 23, 2023 @02:53AM (#34486491)
Wind requires huge quantities of rare earth metals and/or copper. There's no free lunch here.

Rare earths: https://www.bbc.com/future/article/20150402-the-worst-place-...

Copper: https://www.theguardian.com/us-news/2021/nov/09/copper-minin...

Mining both have a devastating impact on their local environments.

Re:NRC Certifies First U.S. Small Modular Reactor Design by moloch-hai on Monday January 23, 2023 @02:12PM (#34493488)
Wind power generation does not, in fact, require any rare-earths at all. (And, they are anyway not rare.) You have asserted this falsehood, and been corrected, many times before. Please do not repeat this falsehood again.

Copper is used heavily in all electric power systems, so does not count against any. Likewise, all industrial energy infrastructures depend heavily on mining, less-electric ones most of all. So, please do not throw up this irrelevant canard again.

Re:NRC Certifies First U.S. Small Modular Reactor Design by moloch-hai on Saturday January 21, 2023 @04:55PM (#34471003)
Expectation is that Indian Point will cost $1B, not counting finding a home for the on-site waste.
Re:NRC Certifies First U.S. Small Modular Reactor Design by arcticbull on Monday January 23, 2023 @02:55AM (#34486497)
First of all you're double-counting decommissioning costs because as I said operators are collecting decommissioning costs during operation. So that XXc/kWh includes the cost of decommissioning but you're talking about it like it's a surprise on top after the fact. It's not.

While you are actually underestimating the cost of Indian Point decommissioning (the scale of which was a surprise to me too - but likely attributable to there being 3 reactors), it has a $2.4B fund that's already been collected and held in trust for its clean-up. But of course Indian Point is in New York, home of the most expensive subway track on earth - $3.9B per mile for the Q line extension. So decommissioning a 67 year old 3-reactor NPP costs the same as 0.58 miles of Q line extension.

Since commissioning, it generated 777TWh based on a 73% lifetime capacity factor (src: Wikipedia). Decommissioning cost $0.0029/kWh generated. Why are we still pretending this is expensive?

On the other hand you know who isn't responsible for collecting the cost of recycling during operation? Solar and wind operators.

I don't know why people saying that nuclear has a waste problem when even cursory research will show you it simply doesn't due to the sheer energy density of the fuel. Further, fast neutron reactors produce two orders of magnitude less waste, and with incredibly short half lives.

The reality is we've been okay with the quantity of waste in large part because the input uranium is just so cheap it hasn't been economically effective to reprocess and reuse it, and nuclear has been such a whipping boy of the so-called environmentalist movement that we just haven't meaningfully invested in developing the technology in decades. The outcome of this so-called environmentalism in the 60s though was 60 years of coal and oil plants.

We've had a renewable, effectively-unlimited source of clean zero-carbon energy for decades now. We've simply elected not to pursue it.

We know the biggest cost of nuclear is the up-front expenditure and there was a really good interview recently with the head of the DOE loan program, Jigar Shah.* He talked about how a ton of the cost was simply that each NPP is a one-off snowflake, and that there were huge savings to be realized by standardizing designs and copy-pasting them onto favorable spots. There are tons of low hanging fruit that we can actually invest in addressing that would dramatically reduce the cost. We just lack the motivation. Stop mistaking the current state with the potential.

* I believe this was on Odd Lots.

Re:NRC Certifies First U.S. Small Modular Reactor Design by moloch-hai on Monday January 23, 2023 @02:21PM (#34493604)
Yet, no one has been willing, in decades, to lend any of their own money to building a nuke. What are you missing? That is a rhetorical question; the topic is dead and buried. Good day.
Re:NRC Certifies First U.S. Small Modular Reactor Design by perihelions on Friday January 20, 2023 @03:53PM (#34459775)
- "Have you seen how big the steam turbine is?"

If your intuition was correct, I think we'd see a trend towards much smaller steam turbines with fewer stages. It's a deliberate choice to engineer them at the size scale they are: the marginal efficiency gains from largest [0], lowest-pressure stages has to justify their cost.

[0] https://power.mhi.com/products/steamturbines/lineup/thermal-... (diagram showing relative sizes of HP / IP / LP turbine stages)