1,482 karma · joined June 18, 2007
---
Avoiding car accident and not smoking is way, way easier than avoiding most effects of a major nuclear accident (fine dangerous and very durable dust disseminated on a vast geographical zone, thanks to wind and rain).
The total amount of victims of the Chernobyl accident is a matter of debate: https://en.wikipedia.org/wiki/Chernobyl:_Consequences_of_the...
I fail to understand which of the challenges I described it may relieve.
> The answer to generic defect is not to have a bunch of incompatible stuff with no commonality.
Having all brands of SMRs build interoperable parts is AFAIK beyond the most wild utopia. The very first step (convincing them to adopt the same architecture) is out of reach.
Yes, the NIMBY effect can kill literally anything, however it IMHO is way more powerful against nuclear than against solar and even wind energy, and as those last expand more and more people will understand that they don't have, in order to obtain electricity, to accept any nuclear-induced risk.
> The PWR is the problem
Which architecture seems more appropriate?
> the US, has unusable regulation, and Europe has wildly all over the place regulation, the needed money is just almost impossible to happen.
This is indeed a real challenge, however it mainly is because of past nuclear incidents and accidents.
> Alvin M. Weinberg was even more right
About the "Faustian bargain"?
> An France did maybe the smart decision in their history when they embraced that philosophy. Sadly they only did so for 1-2 decades and then the next generation came in, was convinced that all those old people were idiots and now that the problem was solved for them they no longer had to pay attention to energy policy.
As a French I doubt so. Details: https://sites.google.com/view/electricitedefrance/messmer-pl...
China invests way, way more on renewable than on nuclear.
Experts (International Energy Agency, McKinsey, etc.) are clear: renewables will produce the majority of electricity by 2050.
https://www.iea.org/commentaries/tripling-renewable-power-ca...
https://www.mckinsey.com/industries/energy-and-materials/our...
This is inevitable (no other possibility is realistic), and this applies to France, where RTE (EDF's subsidiary that manages the electricity grid) has emphasized that 100% nuclear power is out of the question, and that more than 60% nuclear power would be utopian.
This trend is already clearly visible and welcomed by investors.
https://ourworldindata.org/grapher/electricity-fossil-renewa...
The cost of producing renewable electricity is increasingly lower than that of nuclear power.
https://www.lazard.com/research-insights/levelized-cost-of-e...
During their operation, they do not consume fuel, produce waste, or expose people to a major accident, and are therefore and will be preferred.
https://www.bloomberg.com/news/articles/2024-04-09/european-...
The surplus electricity sometimes produced by renewables will be leveraged by increasing storage capacity (electric vehicles, hydro pumped storage, etc.) as well as green hydrogen. It will gradually replace nuclear electricity, which is supposed to compensate for their variability, especially since the latter cannot adjust its production quickly and frequently enough ( https://news.ycombinator.com/item?id=41796580 ).
All of this will reduce the nuclear load factor and therefore increase its production cost (it is only profitable with a high load factor), with strong feedback (the higher the load factor, the more likely it will be).
This will increasingly favor the deployment of a system based primarily on renewables, meaning that nuclear electricity will become less and less useful and more and more expensive, making its fate (and therefore the profitability of the heavy investments required for its infrastructure) easy to predict.
A renewable-nuclear mix will doom the latter.
Yes, running at full-tilt 24/7 is way easier for a nuclear reactor than doing the load-following game (on every account: total cost, maintenance, risk...). They are built for it!
Hydro: yes, to an extent because load-following hydro is mainly done by pumped-storage, many dams are of the "run-of-the-river" type and cannot always load-follow: if there is no incoming (upstream) water they cannot produce any power.
https://www.sciencedirect.com/science/article/pii/S136403211...
https://www.sciencedirect.com/science/article/abs/pii/S09601...
https://www.sciencedirect.com/science/article/pii/S096014811...
Moreover nuclear's aptitude to load-following is vastly over-stated because it has too much inertia (even hot). Even France (shock-full of reactors) always needs to produce approximately 10% ( https://ourworldindata.org/explorers/energy?Metric=Share+of+... ) of its juice thanks to fossil fuel, for a non-negligible part ensuring load-following.
Case in point: France. A household consumes an average of 14 kWh of electricity per day. The capacity of electric cars will exceed 500 GWh before 2035 and 2000 GWh between 2040 and 2050.
Trucks, utility vehicles, and stationary batteries (domestic and industrial) will add to this. Batteries from retired vehicles will increasingly be converted into static batteries before being recycled (see "Redwood Materials" in the US).
In California, when the sun is at its peak (midday), solar power produces up to three-quarters of the electricity. Batteries are charged in the afternoon, when solar electricity is cheap, and released in the evening, when Californians return home. At their peak consumption, around 8 p.m., batteries can supply up to 30% of the state's electricity.
Financially, SMRs are efficient when they are mass-produced and then installed as is, which is difficult to imagine today given the abundance of specific requirements from national safety authorities and site-specific characteristics impacting the installation method. Reactors will therefore have to be adapted (before or, worse, after factory manufacture), which greatly reduces the value of mass production.
Furthermore, the underlying industrialization approach standardizes products and thus increases the risk associated with a generic defect: the discovery of a problem could force the rapid shutdown of a large proportion of the (identical) reactors in a fleet.
This necessary industrialization, and therefore mass production, makes it difficult to claim to only satisfy niche markets.
Even if the SMR becomes a reality, the NIMBY effect alone could wipe it out.
On the ground today, no SMR model is in operation in the West, not even at the industrial prototype stage. Russia has an old, improved military reactor used on a barge (its load factor, as well as that of a recently launched Chinese model, is very poor).
Imitating it would be risky because the total cost of a military reactor (on board a submarine, aircraft carrier, icebreaker, etc.) is much higher than that of an equivalent civilian model. The Navy is willing to pay for features that are decisive for it (long battery life, reduced need for maintenance and surfacing, silence, compactness, etc.) but of no benefit in civilian applications.
Furthermore, a military reactor operates at sea, thus in a huge "cold source" facilitating its cooling, and in the event of an accident, it would likely be submerged far from any populated area. This is difficult to transpose to a national electricity system.
On the ground, the most advanced offering (NuScale) in the most favorable context (the USA) is withering away. Projects in Canada, a nation with expertise in nuclear power, are struggling to get off the ground. In Europe, Naarea, Newcleo, and Jimmy are reeling.
There's nothing new here, as these vain hopes correspond to what Admiral H. Rickover described as early as 1953...
So what? Something which hasn't been done yet must not be attempted? Or is it doomed to fail?
The transition of many electrical systems is a work-in-progress. A complete re-haul of such heavy industry branch cannot be quickly completed, especially during a global crisis.
> You can safely ignore all the households, they barely use any power compared to commercial and industrial facilities.
Nope.
USA: Residential customers (139.894 million) directly consumed 1,509.23 TWh, or 35.23% of the total.
Source: https://en.wikipedia.org/wiki/Electricity_sector_of_the_Unit...
> What is the office tower going to do, use backup batteries from 500 cars in the basement to run dozens of pumps and fans?
A continental mix of renewables can cope most of the time. The point is the 'backup' (when a geographical zone doesn't produce enough and cannot be helped by another zone): dams, batteries, green hydrogen...
> Supplying 30% of California’s power is not 100% backup of the grid with batteries
This is a work-in-progress. 15 years ago some said that renewables will never be able to generate more than a few percent of the current running on national grid.
> We are a ridiculously long ways away from an exclusively solar + wind + batteries grid.
To each is own opinion.
Production trend: https://ourworldindata.org/grapher/electricity-fossil-renewa...
At continental level "no sun and no wind" is extremely rare and doesn't last.
The nature of nuclear power doesn't make it 100% available (no equipment is), therefore a classic way of presenting the challenge, such as "nuclear power is perfectly controllable and it's necessary" (two lies), is a distortion.
The major intellectual fraud consists of considering the characteristics of a type of energy source (renewable, nuclear, etc.) when all that matters is the adequacy of the electricity system, i.e., to begin with, its ability to meet demand.
In terms of the imperfection of sources and equipment, which prohibits us from always expecting them to be ready to produce, the solution is known, applies to all types of sources and equipment, and is already in place: a production fleet containing a number of units that sufficiently reduces the effect of their individual variability (whatever the cause).
The French nuclear fleet is thus made up of a sufficiently large number of reactors (57 in 2025) to make it unlikely that they will all be shut down simultaneously, and for their combined flexibility to increase its "controllability."
This smooths out the impact of imponderables because it is possible to approximate the probability of failure of each reactor (its reliability) and because they are not identical, so the discovery of a defect does not necessarily imply the shutdown of the entire fleet.
A renewable mix at continental level will be way better on nearly all accounts (cheaper, less dangerous, no dependency on any fuel, no durably very dangerous 'hot waste', no weapon proliferation...).
According to the French nuclear industry itself a major accident on one reactor may cost more than 430 billion euro (2013). Source (French): https://www.irsn.fr/savoir-comprendre/crise/cout-economique-...
Biz as usual... https://sites.google.com/view/electricitedefrance/accueil#h....
Case in point: France. A household consumes an average of 14 kWh of electricity per day. The capacity of electric cars will exceed 500 GWh before 2035 and 2000 GWh between 2040 and 2050.
Trucks, utility vehicles, and stationary batteries (domestic and industrial) will add to this. Batteries from retired vehicles will increasingly be converted into static batteries before being recycled (see "Redwood Materials" in the US).
In California, when the sun is at its peak (midday), solar power produces up to three-quarters of the electricity. Batteries are charged in the afternoon, when solar electricity is cheap, and released in the evening, when Californians return home. At their peak consumption, around 8 p.m., batteries can supply up to 30% of the state's electricity.
Is this comparisons solely based upon deaths by (car passenger * distance traveled in NY streets) to deaths by (passenger * distance traveled in the NY subway)? The total amount of car passengers victims of accidents happening on highways, for example, seems not pertinent to me.
We cannot cancel this risk, and we can cancel the risk of nuclear accident by not exploiting nuclear reactor (this is now possible thanks to renewables).
> To date, nuclear energy has > while our renewable experiment
The LCOE is the gold standard.
Comparing an existing fleet of reactors with many hidden costs (indirectly paid for by the taxpayer or the consumer) with the full cost of renewables, and neglecting the cost of any nuclear mishap (accident, waste, decommission...) is a classic trick. In France some even compare the official production cost of the amortized fleet (w/o the investment) to the complete cost of renewables. Yay!
> once it is buried
Who will bury an industrial nuclear reactor during a major accident, and how will they do it? Where is this even only a plan?
Or is it about building it underground, and what about skyrocketing inspection and maintenance costs? Where is this even only a plan? Do your really believe that a broken nuclear reactor vessel vomiting corium will be safe underground, and in such a case why are waste long-term repositories (way less 'active') so difficult and expensive to design and build (as already stated: https://news.ycombinator.com/item?id=44517316 )?
This view is much easier to defend than the one in favor of new reactor designs. However, the fundamental debate is about renewables vs. nuclear, and factually, nuclear, even in a nation with good uranium deposits + good track record + design safer than most other ones... is increasingly difficult to defend.
> don't modernize it too much
The price to pay for this is high because it means giving up on trying to resolve some of the problems of nuclear power (risk of accidents, dependence on uranium, waste, proliferation of weapons, etc.).
> it might still be cheap
Projects over the past two decades show that even projects with improved old reactor designs (rather than radically new concepts) fail.
> Decouple Pod
It seems to me that he misses some key points, for example won't such an underground setup be way more difficult to inspect and even more to maintain/repair? Long-term waste repositories projects are way less demanding and their costs are exploding.
Let's check!
Hitachi Nuclear, hard hit by the reactor shutdown in Japan following Fukushima, canceled projects, even in Europe (inherited through the acquisition of Horizon Nuclear Power) and even their R&D for Generation 3 (ABWR).
Their overall move is moving them away from the energy sector (abandoning their wind turbine division in 2020).
The deal with GE ("GE Hitachi Nuclear Energy") dates back to 2007 and has produced nothing to date: these large industrial companies (inertia...) with very different and poorly complementary cultures have not worked together, and none will be permanently based on the soil of their first customer (Canada).
The combination of their resources and weak hopes will accentuate the difficulty of orchestrating a large two-headed project: Areva NP (French+German), worse.
The BWRX design is truly innovative and very recent; no examples exist.
Specification adjustments demanded on the fly by the safety authority will likely rain down, delaying and increasing the cost of the project.
The client will be sensitive to this, having already suffered this (at Darlington, the host plant itself) and is experienced (Darlington has been in operation for 30 years).
The customer is strict (their nuclear power is renowned for its good performance), well-positioned to benefit from the advice of US and French safety agencies (whose relevant institutes are interested in SMRs), and perhaps a little bitter about not being able to deploy their national CANDU.
This SMR model has been optimized to be low-cost (series effect), which could (the magnitude of the side effects of certain modifications) make it difficult to adapt despite its modularity.
I see a recipe for disaster likely to make the 4 EPR projects and Vogtle look like resounding successes.
There is nothing we can do about it, therefore comparing this risk the the risk induced by nuclear reactors seems moot to me as we can decide to prefer renewables upon nuclear.
> cheap
Nuclear-generated electricity is way more expensive than renewables', and the gap is widening. Source: LCOE (the gold standard) https://en.wikipedia.org/wiki/Cost_of_electricity_by_source#...
> reliable
A continental fleet of a renewables's mix is at least as reliable.
> zero-emission
No, the total lifecycle emissions of nuclear (industrial PWR) is low (10-15 g eqCO2/KWH) but not zero.
> electricity available 24/7
A continental fleet of a renewables's mix with storage (vehicle batteries thru V2Gn, green hydrogen, hydro...).
In order to generate electricity even France burns non-negligible amounts of fossil fuel since the inception of its nuclear fleet: https://ourworldindata.org/explorers/energy?Metric=Share+of+...
> More people will die from plane crashes
One can decide whether he will (or not) hop on a plane. A nuclear reactor and its waste threatens everyone, even very remotely and in a distant future.
Note: my own brother was killed during a jetliner crash (Swissair SR111, 1998).
No. The load factor of the pair of EPRs built in China (5 years late and 60% above the budget) at Taishan is quite bad (.55 and .76).
In France the EPR isn't even producing electricity, while it was to be delivered in 2012 (budget 3.3 billions €, real cost > 23.7 billions €)
> delay and cost overruns caused by local political opposition and lack of vision
Source? An official report (dubbed the "Folz report") explains why the EPR project in France (Flamanville) was a failure, I cannot find "local political opposition" among the causes.
French ahead: https://www.assemblee-nationale.fr/dyn/media/organes-parleme...
> In France all historical reactors worked so well that we did not feel the need to build more.
The context was quite different: https://sites.google.com/view/electricitedefrance/messmer-pl... ... and the real total cost of this "nuclearization" is already huge.
> This lead to talented engineers going to retirement without having a chance to pass on their knowledge and experience
The Civaux-2 reactor was delivered in 1999.
In 2000 the French nuclear sector (at the time "Areva") was trying to sell EPRs (even in France).
In 2003 Finland ordered an EPR and work began in 2005.
How exactly are we supposed to believe that all knowledge vanished, without anyone in the industry to act accordingly, especially while the existing French fleet of reactors (56 at the time) had to be maintained?
> our decision to put it aside for a while
Cause: oil counter-shock (~1985), which (sadly) reducing electricity competitivity https://ourworldindata.org/grapher/electricity-generation?ta...
Even EDF, as early as 1986, considered the nuclear fleet too large: "We will have two to four too many nuclear reactors by 1990," ( https://www.lemonde.fr/archives/article/1986/01/17/nous-auro... ) and this was confirmed by the 1989 Rouvillois-Guillaume-Pellat report. The reason is well known: after the oil price shock, hydrocarbon prices had fallen significantly and sustainably, and they were competing with electricity.
However, reactors were built until the end of the 1990s. Three of them were started after 1985, and four were built in the 1990s. Some were ready to go in 1999 but did only diverge the generate electricity in 2002...
> certified for a lifespan of 60 years
Subject to a successful technical in-depth inspection every 10 years.
An accused defends himself badly by declaring to the judge "I am not the only culprit of homicide!".
Plate tectonic and sismotectonic are also sources of concern: https://link.springer.com/article/10.1007/s10706-005-1148-4 https://www.sciencedirect.com/science/article/pii/S004896971...
For nations devoid of uranium reserves and not absolutely sure to always be able to secure uranium supply (i.e. not a superpower) recycling is an interesting way.
Case in point: France.
Some species of bacteria needed for the vital nitrogen-cycle thrive inside clay.
That's the reason why clay balls/pebbles/pellets are omnipresent in many types of plant cultivation projects.
Stating it is useful because many neglect (or maybe even ignore) this fact.
The way the War Nerd puts it is IMHO the best.
This is a very important rule stated by the War Nerd: 'Most people are not rational, they are TRIBAL: "my gang yay, your gang boo!" It really is that simple. The rest is cosmetics.'
A small human group is compatible with this tribal behavior because the bulk of actions (or at worst their effects) are quickly perceptible to everyone. The larger the group, the less each person understands what is happening, even the effects of what he does.
IMHO it is the genuine and most adequate 'community', because the less abusive way to decide is subsidiarity ( https://en.wikipedia.org/wiki/Subsidiarity ). Constraining people who live near the potential site makes them at best suspicious, they will better memorize this constraint than any benefit from other projects.