Hacker News Reader

Comment by ps4fanboy | Hacker News Reader

Nuclear Power, until those who care about the environment accept that the only clean at scale power solution for base load is nuclear power we will get no where. "Renewables" are great and should be in the mix but they are not an at scale solution.

http://www3.epa.gov/climatechange/images/ghgemissions/source...

By going mostly electric for our transportation and nuclear for power generation we could really make a dent in 58% of this pie.

There are people who simultaneous worry a tremendous amount about global warming who will actively protest against nuclear power.

All while an extremely large amount of pollution is being burned all around us. Be reasonable we need a now solution or there wont be a when.

View on HN

This plan seems to ignore the growth curves of renewables.

By the time any nuclear power expansion plan gets into place exponential growth in photovoltaics and other renewable technologies will have taken over.

Nuclear power would have been a great idea, if we had started at least 10 or 20 years ago. It is unfortunate "environmentalists" did such a disservice to the environment, but we're long past the point of Nuclear being a viable solution.

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

Doubtful that will be anytime soon and we'll most likely always need some sort of baseline power system that's not a from a renewable source.

That ignores the price curve of batteries. Right now, wholesale prices for lithium ion batteries are $150/kWh of capacity, which translates roughly to $0.15/kWh of stored energy. This is less than the price swing in spot prices for energy currently, so we'll start seeing this market expand drastically.

And as electric car batteries lose capacity and need to be replaced, those batteries will still have plenty of storage left for use at home or on the grid.

When batteries get down to $75-$100/kWh capacity, storage will be able to take over much more of the grid. And there are lots of promising flow batteries that can beat lithium ion on price, when it comes to grid-level storage.

Not to mention other energy storage options - of which there are many.

If renewables had had a small percentage of the money thrown at nukes in subsidies and sponsorship, we wouldn't be having this conversation.

1) I think you're off by about 2x for wholesale lithium battery prices; they are closer to 300 USD/kW-hr.

2) a) The economics of battery energy storage are effectively a type of arbitrage in time, where you get some fraction of depth-of-discharge over the time in question.

2) b) In the most limiting case, lets assume that battery energy storage is only being used for daily cycling, that the maximum cycle depth is 75%, and that the investor expects to break even in 5 years. In order to be viable at that breakeven time, the price of energy needs to swing by over 0.22 USD/kW-hr over the day (RMS !). Considering that bulk energy is only 0.05 USD/kW-hr now (average!), that's a huge amount of volatility that needs to be exposed to the market.

I suspect that if customers are exposed to that volatility instead, that we will see enough voluntary adjustment on the demand side of the market to prevent 5x swings in price. With that much volatility available, consumers will demand the savings that aggressive time-of-use pricing offers in such a market.

On the other hand, the economics become much more favorable for energy storage in the frequency support market. For example, the market for primary frequency reserve (PFR) operates on 15 minute time scales and faster. Therefore, the time between battery cycles is two orders of magnitude smaller than the bulk storage market. Right now, there is enough PFR on the US grid (via coal-fired steam and gas-turbine power) that the price for PFR is very low. But as renewables continue to displace fossil fuels as a fraction of total power, the price for PFR will start to creep up.

Agreed, storage is arbitrage, but arbitrage that's going to have a ton of economic purpose and necessity, not just somebody being a middleman.

The Chevy Bolt's batteries have been widely reported to cost $145/kWh, see for example [1]. I'm not sure what's excluded from the whole component to get to the price, but it's a point of reference at least.

The full life of various lithium ion batteries is still being studied and will vary with chemistry, but roughly a partial cycle seems to count as a partial cycle in terms of battery life. So if you're only cycling 75% capacity on a day, then you'll make up the extra 25% in longer life. Most studies I've read of lithium ion battery life but the number of cycles at 1000 +/- 200, to maintain "most" of the battery's capacity, which is why I used the conversion factor of 1000. I would believe a 25%-40% swing in practice, in either direction.

Peaker power plants quite frequently operate at over $0.30/kWh, more than 6x the cost of bulk energy. Energy storage is already gaining in this space [2]. Of course, consumers never see $0.05/kWh pricing, averaging $0.12/kWh nationwide, which is why residential solar makes so much sense for so many people. Consumers that choose demand-based pricing are already exposed to price swings on the order of $0.20/kWh in time-of-use plans both on Southern California Edison [3] and PG&E [4].

The speed with which solar and batteries are changing is going to catch a lot of people in the energy industry completely flat-footed.

[1] http://cleantechnica.com/2015/10/05/chevy-bolt-battery-cells...

[2] http://www.bloomberg.com/news/articles/2015-12-22/batteries-...

[3] https://www.sce.com/wps/portal/home/residential/rates/Time-O...

[4] http://www.pge.com/en/myhome/saveenergymoney/plans/tou/index...?

With regards to cycle life, the amount of cycle life you get by reducing the depth of discharge is very nonlinear. Some internal studies show it be a strong power ((1/fractional_depth)^4 or more).

Nuclear can still catch up.

Photovoltaics have a cap even if the panels are produced for free, simply based on land cost and infrastructure cost. That is also because they are not matchable to the demand curve. As a batch technology (like charging batteries for new electric cars, or powering irrigation) they are great. It for regular on-demand electric use it will always remain an in-fill.

We've had this discussion before, but nuclear is already more expensive than renewables, and getting more expensive. Renewables are getting cheaper.

In addition, nuclear takes longer to setup, and the most feasible solutions being considered that are supposed to solve nuclear's current problems are all in the prototype stage at best.

This neglects how the great expense of nuclear and the long amount of time it takes to build is at least in part due to decades of anti-nuclear propaganda. There are certain activist groups who have created this problem and are now disingenuously pointing to the problem as if they had nothing to do with it.

No, even the French nuclear build showed negative learning effects that had nothing to do with anti-nuclear propaganda. A global nuclear build is going to do much, much worse than France did:

http://www.sciencedirect.com/science/article/pii/S0301421510...

Abstract The paper reviews the history and the economics of the French PWR program, which is arguably the most successful nuclear-scale up experience in an industrialized country. Key to this success was a unique institutional framework that allowed for centralized decision making, a high degree of standardization, and regulatory stability, epitomized by comparatively short reactor construction times.

Drawing on largely unknown public records, the paper reveals for the first time both absolute as well as yearly and specific reactor costs and their evolution over time. Its most significant finding is that even this most successful nuclear scale-up was characterized by a substantial escalation of real-term construction costs. Conversely, operating costs have remained remarkably flat, despite lowered load factors resulting from the need for load modulation in a system where base-load nuclear power plants supply three quarters of electricity.

The French nuclear case illustrates the perils of the assumption of robust learning effects resulting in lowered costs over time in the scale-up of large-scale, complex new energy supply technologies. The uncertainties in anticipated learning effects of new technologies might be much larger that often assumed, including also cases of “negative learning” in which specific costs increase rather than decrease with accumulated experience.

Interesting! The article is paywalled; I don't suppose you could summarize what negative learning effects they found, and the explanations they gave?

If we can go mostly electric for our transportation, as you suggest, that means we have cheap and plentiful battery storage. If we can easily store power, there's not much advantage for using nuclear power over wind and solar.

Not necessarily true. Liquid fuels are used for transportation but only for niche generation, because they have higher costs, both in capital equipment and in fuel cost.

Likewise, battery cost for transport can make it attractive, but that doesn't necessarily make it better for stationary generation. Scale beats everything when it comes to generation.

Scale matters for massive thermodynamic processes for effieiceny, but 21st century technology has become much more decentralized. You don't need massive installations of solar or wind at one location, the can be distributed throughout the grid. Massive scale is possible, but it's not required once you get away from heat differentials as the core of your energy process.

It still matters for transmission, construction and maintenance costs, as well as resources like land and labor.

You may not need scaled installations, but they will still beat out the alternatives in any industry shakeout over the long term. We build large scale factories, farms, ports, airports, roads and retail outlets. These happen because they work. Even if the energy production at scale gave no advantage, all the other advantages of scale means that it wins out in the end.

> the only clean at scale power solution for peak load is nuclear power

Nuclear is fantastic for base load, but peak load? Steam bypassing costs as much as generating electricity does, but with nobody wanting to pay for power they didn't use. Technically a good solution, but falls apart economically.

Sorry I meant base load.

> nuclear power bla bla global warming bla bla

Chernobyl? Fukushima?

The technology has the potential of destroying huge areas for good. Do we want that?

A complex system that is attacked on multiple threat vectors is likely to fail - no matter how many redundandancies are put into safety. I can tell you that from looking at CS/network security.

A complete failure/breach of a big reactor is very costly.

> destroying huge areas for good

Even Chernobyl isn't destroyed for good, though it's still an expensive mess to clean up. Fortunately, nobody builds reactors with such an idiotic design (a positive void coefficient[1]).

The actual site at Fukushima Daiichi is going to to be a long, expensive cleanup, but that's about the extent of the damage. Concerns about the surrounding area is just radiophobia and an extreme overreaction by the Japanese government. With how incredibly destructive the tsunami was, Fukushima Daini (which was also badly damage by the tsunami) showed us how incredibly safe nuclear power can be.

> can tell you that from looking at CS

More CS people need to study the lesson of incidents like therac-25[2]. It's very important to learn what fail safe means, which is very possible to implement, usually by avoiding as much of that "complex system". I trust the inherent and engineered safety systems of a nuclear reactor much more than I would trust any software project - including my own.

> A complete failure/breach of a big reactor is very costly.

So are fly ash slurry spills. You're holding coal to the same standards, right?

[1] http://www.hiroshimasyndrome.com/chernobyl.html

[2] http://sunnyday.mit.edu/papers/therac.pdf

Caesium 137 has a fucking halflife of 30.7 years. There is still a lot of that material in Ukraine and Belorussia. The economy and agriculture of both countries sustained severe damage as a result http://www.greenfacts.org/en/chernobyl/l-3/5-social-economic...

now were just very lucky that there was no full meltdown of the reactor core. Can you claim without flinching that this will never ever happen? I doubt it.

Just look up 'heroes of Chernobyl' - we were very close http://knowledgenuts.com/2014/04/13/when-three-divers-swam-i...

The fact that an accident of similar magnitude happened after Chernobyl sort of invalidates the point that it was all exclusively due to faulty reactor design.

In Germany this assessment led to a full shutdown of all nuclear power plants.

> There is still a lot of that material

I never said there wasn't. The cleanup is ongoing and will take a very long time, but it still isn't "destroyed for good". At least the caesium eventually goes away. The Pb, Hg, and As in the 1.1 billion gallons of fly ash slurry that the Tennessee Valley Authority spilled[0] over homes and into the Emory and Clinch rivers doesn't have a half life.

> economy and agriculture

The tsunami also did incredible damage to the economy and agriculture of Japan.

> Can you claim without flinching that this will never ever happen?

Absolutely. Even the idiotic RBMK design can be run safely after being modified to fix the void coefficient. I suggest actually reading my link[1] above, which explain in great detail the problems in the design of that reactor. Even with those problems, it took incredible effort[2] to trigger the steam explosion.

> 'heroes of Chernobyl'

I'm give a lot of respect and thanks to those that had to deal with Chernobyl's immediate aftermath and the 600,000 liquidators that were part of the cleanup afterwords. In particular, the 3828 people that cleaned up the worst section of roof in two minute shifts. Their story - in their own words[3] - is something I will never forget.

[0] https://en.wikipedia.org/wiki/Kingston_Fossil_Plant_coal_fly...

[1] http://www.hiroshimasyndrome.com/chernobyl.html

[2] https://leatherbarrowa.exposure.co/chernobyl

[3] https://www.youtube.com/watch?v=FfDa8tR25dk

Tsunami and coal dusts are less severe than radiation hazards. I think this comparison is deliberately misleading.

You're going to have to show some proof for this claim, because you seem to have the risk assessment backwards.

> Tsunami

The tsunami in Japan killed over 15,000 people (with several thousand more missing, likely dead). Radiation from Fukushima Daiichi (or anywhere else) hasn't killed anybody yet, though a handful of unfortunate workers and others in the immediate vicinity may have cancer problems in the future. Hopefully those cases will be treatable, but we won't know for some time.

15,000+ dead seems to indicate some tsunamis are far more dangerous than nuclear power.

> coal dusts

You might want to look again at the contents of fly ash. It's radioactive, full of a long list of heavy metals, with trace amounts of nasty things like dioxins.

Why? The tsunami killed more people than the radiation is estimated to even with the most generous estimates of how many people will get sick from radiation. Hell, even the stress/evacuation caused by the worry of radiation is likely to kill more people than the radiation itself would have if they didn't take those steps.

http://www.nytimes.com/2015/09/22/science/when-radiation-isn...

The former head of the IAEC has come out on record against this notion. She said that the paltry few nuclear power plants in development are no where near far along enough to make a difference for over 20 years from now, while renewables are delivering, albeit still too slowly, right now.

Earth has a set amount of uranium that is destroyed when creating energy.

Currently we have about 200 years of uranium available for economical extraction. But if we tried to replace all fossil fuels with nuclear, we would be out within a few decades. Then we'd need to move to other non-renewable materials (thorium?) or start mining astroids.

http://www.scientificamerican.com/article/how-long-will-glob...

And the mining of uranium looks awfully similar to the mining of coal. (Of course you need far fewer uranium mines for the same amount of energy.)

https://en.wikipedia.org/wiki/Uranium_mining#/media/File:Ara...

Pick a problem, you cant solve them all at once.

- mining is pretty much a necessity of life at this point.

- 200 years to come up with solutions to energy production while not turning the planet into an oven sounds pretty good.

You've misunderstood my comment.

We have 200 years of uranium left at current rates of consumption. But if we replaced all fossil fuels with nuclear, we'd consume 10 times more per year and we'd only have about 20 years of energy available before needing to move on to a new energy source.

Uranium exploration is pretty much non existent these days, I would'nt be comfortable saying how much there actually is. Peak oil has been coming for decades but they keep finding it.

Okay, it is wikipedia, but:

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

"Adherents claim that with seawater uranium extraction, there would be enough fuel for breeder reactors to satisfy our energy needs for 5 billion years at 1983's total energy consumption rate, thus making nuclear energy effectively a renewable energy."

Yeah, breeder reactors are even mentioned in the sciam article that nostromo posted. He's cherry-picking his info.

There are only four breeder reactors in operation, we'd need a lot more rather quickly for them to be useful.

Also as with all things nuclear there are an awful lot of claims and not a lot of known things here. Nuclear somehow always looks much better in theory than it does in practice.

There aren't a lot of problems though, there is just one: We rely on fossil fuels. Switching to nuclear fission won't change that in the slightest.

Also even if you could get sufficient agreement on the need for more nuclear power plants - which you won't - it would still take way to long to build enough of them to have an impact.

It makes much more sense to invest in renewable energy, the energy grid and research into nuclear fusion.

With breeder reactors, the Uranium should last much longer. I read somewhere that you get approximately 100 times as much energy out of the same amount of Uranium when you use a breeder reactor, and don't throw the fission byproducts.

So assuming we have 20 years of (currently discovered) Uranium, breeder reactors should make that last a theoretical 2000 years at current energy consumption levels.

> Currently we have about 200 years of uranium available for economical extraction.

I responded to your claim previously[0]:

> In that 200 years a whole lot more uranium will become available for economical extraction, exactly the same as what happened with oil and natural gas.

Do you have reason to believe this will not happen?

[0]https://news.ycombinator.com/item?id=10623187

I'm no expert here. I'm simply parroting what the scientists at the Nuclear Energy Agency have reported.

OK, but remember that the experts said the same thing about oil and natural gas, and were wrong in that case. Here the estimate of ~10 megatonnes assumes that seawater extraction will never be economically viable, one possible source of error.

If you're interested in the history of "peak resources" here is a good book about Paul Ehrlich's famous bet with Julian L. Simon:

http://www.amazon.com/The-Bet-Ehrlich-Julian-Earth%C2%92s/dp...

> Do you have reason to believe this will not happen?

What's kind of amazing is that people don't believe that we will develop a solar powered technique to extract carbon the from the atmosphere and create liquid fuels.

Like an artificial plant? If it is possible to engineer something that works better (and there is no reason to believe it is impossible) that would be a huge event.

You may be interest in:

"A New Dawn for Industrial Photosynthesis"

http://link.springer.com/content/pdf/10.1007%2Fs11120-011-96...

By 200 years we will probably have fusion working anyway. Problem mostly solved.

I think in your link "economical extraction" means "economical extraction at current or similar prices". At suitably high prices, even extraction from seawater becomes reasonable (google gave me http://www.technologyreview.com/news/514751/novel-material-s... with an estimate of $1e3–2e3/kg). I assume this would have a limited effect on electricity prices because uranium prices are probably a minor component of nuclear power costs.

This is the same argument that the proponents of peak oil put forth. Discovered uranium deposits != all uranium deposits on earth. When price pressure is put on uranium, we'll have more speculators looking for it. (Which the article addresses.)

Also, Economically unaccessible today != economically unaccessible tomorrow. Our mining technology gets better all the time, especially if there's price pressure on it. We're accessing deep-sea oil deposits that we weren't able to twenty years ago. (Which the article addresses, but you omitted the key point!!! We have a 60,000 year supply of uranium in the oceans!)

The chief problem with coal isn't pitmining, it's the pollution from burning the stuff.

Don't fall for the mistake of taking known reserves to be the actual amount in the earths crust.

'Oil is about to run out' predictions have been falling for this for a century now, and continue to make the same error.

There is no economic reason to find resources out beyond a set timeline. So they aren't looking for them, and so the reserves always look like they are going to run out in 30 years.

The actual environmental impact of mines is negligible over the long term, provided the mining is done in a country with good environmental protection in place.

> 200 years of uranium

As other people have mentioned, that's known 235U.

We have enough domestic 232Th to satisfy our energy needs for a very long time. If we use breeder reactors that can utilize 238U, we also have sufficient availability for well-beyond any reasonably foreseeable future ("many millenia").

Besides, that gives us plenty of time to actually develop other methods of providing energy. Right now every day we fail to deploy nuclear is another day of coal/gas. Even if we used your silly figure of 200 years, it shouldn't take that long to develop a proper solar+batteries replacement (heck, we should have fusion before the 200 years is up).

> mining

There are rare earth mines that we aren't operating right now because it's too expensive to dispose of the Th. In any serious nuclear deployment, the fuel itself would be free (or nearly free).

> far fewer

Comparing U or Th mining to coal mining is silly when you only need to do over 2,000,000x more coal mining.

And so we build nuclear power plants in Nigeria (population 175 million)?

Only if you agree, that you move into a unpopulated area, that the nuclear power plants are built in your garden, that everything is paid by you and you also take care of the waste. Thank you.

I'd be happy to live near a nuclear plant and waste disposal. Why not? It sure would beat living downwind of a coal power plant like many of us do now.

It is a well-known and fully solved problem of how to deal with nuclear waste. The reason it is not dealt with is purely political, at least in the United States: an anti-science Senator who was pandering to anti-science constituents in his state by killing the waste disposal site.

Lots of people keep saying this, but I really don't think it's true. Solar and wind work fine at scale. There is nothing un-scalable about them. What they don't do is work at night, and in that case nuclear is a fine option. Although batteries that can run a house overnight aren't that expensive.

Wind works at night. :)

Given nuclear's huge expense, huge lead times, and intense local political opposition, and comparing it to how battery&renewable is getting much cheaper with time, I don't doubt that battery&renewable will be cheaper at scale than nuclear power will be by the time that either could be built out.

The expense and lead times and nimby-ism which cripples nuclear are mostly regulatory.

If you apply the same constraints to wind generation (which is increasingly happening) it looks even worse than nuclear.

I've never seen convincing numbers that the expense is regulatory. And the nimby-ism is not regulatory in the least.

There are always grand claims of how cheap nuclear can be, but it has never been seen in practice. Sometimes people even claim that nuclear is cheap by only looking at the operating & maintenance costs, ignoring the true costs of nuclear, which are all upfront.

What type of nuclear constraints could be applied to wind generation? Validation of designs? There's very few turbine designs out there. I can't see how wind could be ever be made more expensive than nuclear, but am open to new data.

Where you can put wind power, how many approvals you must have, how long the lead times are.

If you can't see how regulatory process increase costs, use this thought experiment. You want to build a house. You buy some land for it, and submit the plans for approval. Now spend the next ten years paying an architect to complete forms and submitting them for review. You have to pay the carrying cost of the land and the architects salary. The cost of the building materials for house remains the same. Can you now afford the house?

If we went fully "renewable" how much industrial waste would these batteries that everyone needs create? Solar and Wind really doesnt work for everyone, where it is cost effective, available and reliable it makes sense to mix them in, but you cant tell me you fully expect every country to go 100% Solar and Wind?

Global population is concentrated near the equator and near the coast, both great conditions for solar and wind.

Paris is a good example of that.

Denmark is a net power exporter, from renewables.

The industrial waste caused by solar/wind would pale in comparison to the existing waste being generated by electric cars and mobile phones.

And of course you can't expect every country to go fully solar/wind. What you need are integrated, continent wide power grids and alternative energy storage mechanisms like molten sands etc.

Batteries are really not that bad for industrial waste. Lead-acid are terrible, but there are better options when it comes to daily storage of lots of energy.

Compared to coal, give me batteries any day! And we don't really yet know the full effects of all this cheap natural we're getting through fracking. Though I'm sure fracking is probably better than coal mining, we'll have to wait to see.

I don't expect every country to do it this year, or this decade. I also don't expect every country to have nuclear power either, because it's expensive. You think uranium mining is clean? Constructing all that stuff is free? Of course not.

The amount of solar panels needed for a house (or their share, in some solar farm) is miniscule. It's 4-7 panels. If we want to store excess for night, maybe 10. The batteries fairly small (using lithium, a sufficient battery could hang on the wall. Maybe two tesla power walls would do it). The waste isn't that bad (nowhere near the amount of waste people create yearly in plastic or other landfill garbage).

It's totally doable in first world countries. And it would go a long way in developing countries too, for the same reason cell phones gained popularity faster than landlines, because they require less infrastructure.

Any idea what proportion of Western Europe would need to be covered in solar panels and wind farms to reliably give 90% of the required kWh per day? The other 10% could come from tidal. But you'd need much of the potential hydro capacity to act as energy storage. (Assuming battery/hydro back up on a daily basis).

There's a proposed roadmap for much of the world here:

http://web.stanford.edu/group/efmh/jacobson/Articles/I/Count...

Not that much, actually: http://www.techinsider.io/map-shows-solar-panels-to-power-th...

It certainly seems more feasible than some feeble attempt to cover the Earth with a reflective surface as a geoengineering measure to counteract global warming.

Promising. However naive extrapolation from that source would would be a significant underestimate of required capacity. Western Europe has significant seasonal deviations in sun. Our governments would not accept direct electrical imports from Africa. We also consume much more energy per capita that other regions.

What is expensive is replacing all the current energy infrastructure. Even if you went out and just re-built coal power stations and oil refineries it would cost trillions. We sometimes forget in first world countries that a lot of the reason we are so wealthy is because we have an enormous stock of physical capital paid for over the centuries.

I agree. And the fact that you need to store the energy during night and winter is not a big deal for technology either. We have been dealing with a similar problem with food supply since agriculture. I think it looks so unfeasible, because today people don't even recognize when we have a winter - that's how well we have solved the problem.

Making those shiny new batteries is an extremely dirty and polluting process. Again peopld ignore what they dont see.

> Just keep repeating it, but it isn't true.

Please edit uncivil language like this out of your comments when posting here. Without that (well, that and the uppercase yelling), this comment would be a fine one.

https://news.ycombinator.com/newsguidelines.html

Fixed. Didn't know a single word in caps was considered yelling.

Well, there were two ;) Thanks! That reads much much better.

Reply on news.ycombinator.com