One of the best things about the solar/wind + storage combination is that it's decentralized and cheap at small scale. This makes serious electrification of remote and impoverished places viable. It reminds me of the wildfire spread of cell service in places that didn't have phones before, only on a larger and more important scale. Give a village cheap electricity, and awesome things will happen.
This gets to another point that bugs me in these discussions... the pushback against wind/solar dominance by the pro-nuclear crowd. Insisting that nuclear power is the right way to go is western-centric. It's fine for the US, Europe, Japan, and other advanced nations that have the infrastructure to support it. But is it a solution for Peru? For Somalia? Of course not. Wind/solar/storage, on the other hand, is totally viable as a solution for even the poorest nations. This alone is an argument for solar over nuclear.
I wrote a couple pro-nuclear comments here. I'm perfectly happy with wind and solar and think we should roll it out as fast as we can, as well as nuclear.
Many of the wind/solar advocates, on the other hand, are strongly anti-nuclear. Given the state of the climate that just doesn't make sense to me. All these technologies have their own advantages, and we should use each where it's most effective.
I think there are two separate dynamics here, and its worth keeping them separate.
Firstly, there is the traditional concern of the green movement with nuclear waste and nuclear weapons proliferation. Nuclear advocates say these concerns are misplaced because of nuclear technology innovations (although its not clear how close the new technology is to wide deployment).
Secondly, there is an economic and free-market concern. Nuclear power plants generally only get built in highly regulated, centrally planned electricity markets with a great deal of Government financial support.
The projects to build them, at least in modern Western states, tend to be complex, expensive and prone to overruns. These projects generally need the plants to run all the time with guaranteed rates for the business case to stack up.
This is entirely at odds with solar and wind, which are highly democratised -- your Aunty can put solar on her roof -- and even when deployed at 'grid scale' tend to suck the profits out of wholesale electricity markets, because they have the lowest short-run costs and always out bid other energy sources.
In deregulated electricity markets renewables plus gas beat nuclear on price. We need storage to push gas out of the market.
The only reason Aunty can put solar on her roof is because she has the centralized electrical grid to back her up. Put another way, if Aunty had to choose between roof solar and centralized grid (exclusive of one another), most Aunties will choose the grid because it is much more reliable.
That's how we ended up with centralized generation and a grid in the first place--the initial roll out of electricity was highly localized, with factories, buildings, and blocks each having their own generators. Centralized + grid beat that architecture on reliability and cost.
So, the grid needs to be available with or without solar. Thus the "free market" concern is less of a differentiator than it seems on the surface.
This isn't about grid or no grid. We will have a grid for the foreseeable future.
This is a question about how distributed energy generation with battery storage shifts more and more loads to the edges of the grid, and onto private networks.
And then, relating to back to the question of whether nuclear is compatible with renewable energy, the question becomes how does this trend affect what technologies make the most sense?
Big expensive nuclear makes a lot of sense in an environment where all the loads are on are centrally managed and planned grid with either guaranteed rates of return, or steady wholesale electricity prices, but this is not what the future looks like.
My point is that all electricity markets, at least in the U.S., will be highly regulated and centrally planned for the foreseeable future, because customers absolutely expect the grid to meet 100% of their needs no matter what happens.
More generation might move to the edges, but that is simply a factor that will be taken into account by the regulation and central planning.
There has to be central generating to meet demand that local generating fails to meet; but the choice of fuel source doesn't have any implications for "free market" concerns. Electricity is not a free market. For example, net metering is only an option for home solar because a federal law says it must be. In a free market, the central electricity utility could simply decline to purchase from edge generators. Or decline to connect home solar people to the grid at all.
I'm neither pro- nor anti-nuclear. I'm not afraid of it and recognize that it has value (I really like the idea of using it for base load generation with wind/solar), but I also think it's a very limited solution, due to its expense, complexity, and political difficulties. If we can build a purely solar/wind/storage model that is inexpensive, safe, and uncontroversial, I'd much prefer that.
I am not anti-nuclear, I am just anti-nuclear (of the type that is useful for enriching for weapons) for the global solution to electrical power, given that for a general solution you need something that politically, you don't worry too much about your enemies having.
Look at the situation with Iran for instance. Any development of nuclear there is viewed with extreme suspicion, even while we are trying to reduce global oil dependence.
If Iran was rolling out solar and building silicon foundries, we might get into trade disputes over it, but we are not going to be that paranoid about them purifying silicon as it generally does not go boom.
edit - anyone pro-nuclear power who is not pro-nuclear power for Iran should really think on this one. A solution for electricity generation that you are scared of people having is not a good general solution.
edit 2 - I am not meaning to dump on Iran here, is just that the Iranian situation with this has been making the most news recently.
The Iran situation highlights what I said in the first place - not only is nuclear not a technical option for the majority of the world, we would be actively hostile to it for a lot of countries. So basically, the nuclear > solar position is, consciously or unconsciously, relegating parts of the world to permanent poverty, deliberately leaving them behind.
And again, take a remote village. Give them electricity and decent internet. What happens? Their lives get a lot better, their opportunities grow. But focusing on government-centralized systems will continue to exclude them.
I have talked to engineers from oil and gas as well as from nuclear industries about solar. The oil and gas folk generally like it and have an eye on the jobs, whereas the nuclear folk haven't looked into it and claim it is pointless. One of them even thought that the incoming solar is far less than we use and spat his drink out when I told him the actual numbers involved.
Very few people oppose nuclear power for Iran on principle. Opposition to development so far has been based on the obvious evidence that in Iran, "nuclear power" has so far been a fig leaf for developing a nuclear weapon.
The framework that the U.S. is negotiating with Iran right now will allow them to continue nuclear activities that legitimately lead toward peaceful nuclear power.
I'm genuinely wondering, if it is possible to rob 250 million pounds' worth of diamonds from a bank vault in London, how difficult it would be to obtain access to some nuclear waste.
Agreed. The only objection I have to wind and solar is that they are insufficient. We should use them as much as possible, but we're going to need nuclear if we want to replace fossil fuels.
The more advanced (Stanford Energy Symposium) research that I've seen suggests that there is only one possible power source that we can look to long term to meet the energy needs of the world, and that is Solar Energy. Nuclear Energy just won't be possible to build out quickly enough, inexpensively enough - even if we could do it safely.
In particular, check out Nate Lewis's Solar Energy 101 (https://itunes.apple.com/WebObjects/MZStore.woa/wa/viewPodca...) - In the Global Climate and Energy Project (GCEP) where he makes a strong case that the only choice is Solar - not because it's a good one, but because there is nothing else out there that can meet our needs (And he spends a bit of time talking about Nuclear Energy, and why it comes up short).
Incoming solar hitting the ground is ~ 89,000 terawatts
At current conversion efficiencies and accounting for transmission losses, this translates to an area of about twice the size of Portugal spread out globally.
This is an article about energy storage. Nuclear doesn't have much to do with that. Nor does the alleged aversion of 'green people' to nuclear. Please give it a rest.
Human life? Is it? Remember famines that used to regularly wipe out the peasants in most civilizations? It's been a while since disasters on those scales happened in most countries. Climate change doesn't sound so bad when you compare it to normal life a few centuries ago.
yes and no... climate change is not about having +2 degrees globally, we can handle that easily even if sea rises significantly. Not even about extinction of a lot of species, which will trash some foodchains around the world.
It's more about longer perspective - some effects are cascading, and in longer uncontroleld run, we might end up with pretty much inhabitable planet (at least for mankind). Life itself will handle this easily, mass extinctions happened many times in the past for various reasons, mankind in its current level of evolution would be probably over.
For a look at what's actually projected for the climate, read the book Six Degrees by Mark Lynas, who read about 3,000 peer-reviewed papers and summarized them, with extensive references. It's quite frightening, and far beyond anything that human civilization has ever experienced, even at +2C. If we go all the way to +6 it's hard to imagine anyone surviving. Somewhere between +2 and +3, climate feedbacks kick in and take it several degrees further with no more help from us. Right now we're at +0.8.
I didn't say anything about humans, but even the most anthropocentric among us can see that killing everything in sight serves to hasten our own demise. And yes, it is. Read the latest predictions, they are really that bad.
There are other possibilities than nuclear and fossil energy without those vast risks.
The OP article is all about that.
Nuclear is not a viable alternative energy.
Westerners arguing that Nuclear is better for them in no way means solar is not better for an African village in the middle of nowhere....
Or are you arguing that Westerns should adopt solar so that African villagers get slightly cheaper solar? That seems like an inefficient form of aid.
On a side note: solar is not cheap on a small scale, it might be convenient, green, safe, etc. but not cheap (cheap infers low cost versus conventional power generation)
No, I'm not arguing that Westerners should adopt solar as a form of aid for Africa. I don't see how you could have concluded that from what I said.
And narrowly measuring "cheap" in cost per kwh does not address my point - that the up-front cost to get power at all is far cheaper for solar/wind + storage. A village is not a city. It doesn't need a gigantic power plant, and can't afford one. Look at it this way - I could buy myself a can of soda for a buck, or I could buy a bottling plant for a million bucks. In terms of cost-per-liter, the bottling plant is a huge bargain. But I don't need a bottling plant! It's actually useless to me. A can of soda? That's useful, and scaled correctly for my budget and needs.
Good point. What I really wanted to say was "American" in terms of cultural bias, but "developed" in terms of social structure. Got neither right, I suppose.
I argue against wind and wind alone, it is a very unreliable source of energy, I think it is crazy to rely on gusts of wind to power a population. Nuclear is the only technology which offers a base load without burning fossils.
Solar I am happy with, it could be better than nuclear but the storage problem needs solving before it gets onto everyone's roof.
That's another problem with solar, people are putting up fields of the stuff when there is already enough roof space. In my mind large retail shops should be mandated to fill their roofs with solar.
I live in France, where nuclear energy is dominant, and I can tell there are some serious issues with nuclear power.
First, nuclear can be extremely inefficient since it's heavily centralized : I have friends working for energy distributors telling me they can lose more than half the energy between the power plant and the client's house.
More importantly, nuclear waste disposal and power plant cleaning costs are heavily underestimated. I've worked for the CEA, the French nuclear agency where they invented the first French atomic bomb and reactor in the sixties, and some experiments aren't still cleaned up yet in 2010's (the current objective is to fully clean up the laboratory in 2025-2040). That is a massive cost which is not factored into the kW price.
I think decentralized technologies are the way to go regarding energy (using local loops and "smart grids") even if they aren't viable now (solar panels cost more in energy to produce than they give back during exploitation for example).
About: ".. they can lose more than half the energy between the power plant and the client's house." - this seems extremely unlikely to be a representative case. Maybe your friends were telling you about some extreme case (they are always the best stories :)
I think you'd benefit from looking at some actual data. In practice, wind is predictable enough.
You don't need baseload. It's a concept that makes sense when you have power plants with flat output profiles.
What you actually need is to always match the demand. Nuclear isn't some magical silver bullet here. As it turns out, it's simply far too expensive to use for fulfilling the daily peak demands.
Wind + storage, on the other hand, provides base load just fine. You just need to capture enough wind energy over the course of the storage capacity / baseload to keep it stable. On sufficiently large scale, you can get its probabilities up to the level of nuclear plant uptimes.
We're turning a bandwidth problem into a caching problem. That's the game-changer.
If we do get cheap scalable energy storage, it's pretty much game over (in the long run) for everything but solar and wind. We are basically drenched in free energy, but we can't store it cheaply enough. If that's fixed, we're done here.
Nuclear fission may still have on-paper advantages in some markets/climates, but the high PITA (pain in the aXX) factor would probably mean we wouldn't bother going there. We'd just build transmission lines, more solar/wind, and more storage even if it were marginally more expensive just to avoid the headaches of nuclear energy.
Storage helps nuclear too, with less demanding requirements. You can keep running your nuclear plants at full blast and use storage to average the load. If the storage is cheap enough that'll cost less than partially idling the nuclear plant.
Wind and solar, like nuclear, have costs dominated by capital cost, but with less predictable output. To run civilization on those alone, we'd need quite a bit more storage, to cover times when it's cloudy and still for a couple weeks. Long-distance transmission helps but that's not free either.
Particularly now that Gen III+ plants are on the market, fast reactors are maturing (in Russia at least), and half a dozen startups are working on molten salt reactors, I wouldn't take nuclear off the table. We need every non-carbon energy source we can get.
Wind over a large area is far more constant than you might expect the problem is you need a vary large area (east - west coast) to see this. http://earlywarn.blogspot.com/2010/04/averaging-us-east-coas... Still the major advantage to wind is it's the cheapest power out there. (Yes lower than coal, nuclear, hydro, etc.)
PV Solar is also vary steady in the areas you would put it. (AKA not the South Pole.) Storage is useful, but transporting power over distance is much more useful than you might think.
Finally, Hydro has a lot of built in storage allowing you to double output for weeks at a time.
Wind over a large area is far less constant than its enthusiasts claim. Check out the California ISO wind power graphs [1] and the PJM wind power generation graphs.[2] That's real world data over areas over 500 miles across. About 4:1 variation over a day is typical.
To average wind over the entire US, a big network of ultra-high-voltage transmission lines would be needed. This is quite possible; China is building one, with transmission distances up to about 1200 miles. Acquiring the right of way for a UHV line, which has roughly the space requirements of an interstate highway, is the main problem in the US. Long-distance UHV lines are usually DC; the conversion equipment is cheaper than running 3 wires instead of 2. There was talk a few years ago of an "energy corridor" going up through the Texas panhandle and northward, to the good locations for wind and natural gas. The corridor would carry both power lines and pipelines.
Stupid question: if UHV lines need roughly the space of an interstate highway, could the right of way be acquired by building them along interstate highways? There's lots of grassy medians in which to put towers, and the lines themselves are comfortably off the ground.
How much of that observed variation in production is due to the current state of energy demand, and the state of other generators on the grid? It seems that turbines are often powered down.
I wonder what sort of distribution algorithms they use to disperse the wind power. Seems like an appropriate optimization algorithm would do much to solve this issue more than blind infrastructure investment.
There are a surprisingly small number of good wind power sites. California has only four really good onshore sites, all of which have large wind farms in operation. Wind power, like hydroelectric power, is very dependent on geography.
Wind and solar might have large capital costs in-total for similar capacity, but you can deploy it in much smaller increments and time periods than nuclear plants - that goes for both actual deployment as well as R&D investments. I think that makes a huge difference in how the technologies are rolling out in practical market terms. I expect it will be similar for storage technologies - much lower capital investments required for a given increment of technology improvement & roll-out.
Now, new small fission reactor R&D may work out differently, but we have yet to see any of those efforts come out of the gate in terms of being even close to ready-to-install states.
Wind and solar have a fairly beneficial failure scenario (ie nothing happens, save for the few turbines that have destroyed themselves during adverse conditions).
You'll never be able to compete with that, fission, fusion, whatever. We still don't know where to put the waste, we still refuse to acknowledge the ability to reprocess said waste, and so forth. You don't even need more storage to run wind and solar as base load; you simply need to overbuild capacity and maintain a sufficiently intelligent transmission network. The sun is always shining and the wind is always blowing somewhere.
The advanced reactors I mentioned produce about a hundred times less waste, and it's back to the radioactivity of the original ore in a couple centuries. Encase in a block of glass and bury it. They can also use our existing waste stockpiles as fuel, so we'd end up with less waste than we have now. No renewable can accomplish that.
I did see a paper a couple years ago on running a section of the U.S. electric grid on wind/solar alone. They had a computer try about ten thousand scenarios, and found that the cheapest was to overproduce energy by a factor of three, and add a bit of storage. So at least for that region, wind/solar would have to be less than a third the cost of fossil per kWh to be competitive. I certainly hope it gets there.
In the meantime, we can't really afford to be picky about our non-carbon energy sources. Modern nuclear plants have excellent safety records, and more advanced reactors look even better. Let's just not build any more of the 1970's-era plants that have run into trouble. We've learned a lot since then.
In my limited experience [1 windmill that i am currently working on] the biggest issue we are facing is lightning strike. The windmill is located on a small island .6 miles off of the coast of Maine. the structure rises 20+ feet over the tops of the trees. The Seawind is pretty consistent so wind seemed like a good idea, but the first strike fried the innards of the turbine and compromised the lines running to the battery array [the system uses a ruggedized Ranger power management computer to charge the batteries (28 inline 1.5 amp boat batteries) from either the windmill or the diesel generator]. I know very little about all of this, but it has been left to me to get it working. The system only has to power 110v outlets [30+], a water pumping system [well pump + distro pump], and a washer and dryer. As it stands we run the Genny for ~6 hours then use the batteries for ~18 hours. The island is only inhabited for ~2 months all year, but it is still expensive and dirty. The Island is conifers rooted in primeval moss on top of pink granite. there is little to no dirt, and i have limited access to TNT and my biggest rock drill bit only goes down 2 feet, so the grounding system (which i think is where the fault in the system is) is my biggest concern. As of now, i am going to attempt to use a surface, chemical grounding system, but, again, i don't really know what i am doing so it is pin the tail on the donkey with my hands tied behind my back. If anyone has experience with windmills and/or grounding i could desperately use some advice...
You need a lightning protection consultant. Lightning protection is a solved problem - power lines, large communication towers, and wind farms are struck all the time, without serious problems. Doing it right can get expensive, but is quite possible.
Grounding on hard rock is tough. There are methods for measuring ground conductivity, and in a situation like that, you'll need them.
We have a lot of our equipment tested in Florida (where they have some large testing centers) for lightning tolerance. I've been told, that you don't actually verify the equipment can be safely struck by lightning - as anything that is hit by lightning, is immediately destroyed. The objective is to ensure that equipment nearby a lightning strike doesn't suffer secondary damage (possibly from a power surge in the line). We install lightning arrestors near our electronics, but anything that actually gets hit is replaced.
Airplanes though, are supposedly struck by lightning, and continue to work. I'm guessing that this has something do do with them being suspended in air, and not having a connection to ground.
Antenna towers get hit all the time, and the transmitters and receivers survive if installed properly. First, there's a spark gap between the antenna and ground, with big metal contacts, copper or silver, a short distance apart. That diverts most of the lightning bolt to ground. Then the feed line for the antenna has a big inductor, a coil made of heavy busbar, usually in a grounded can. This is often placed through the wall of a grounded metal equipment enclosure.
The inductance blocks a fast risetime lightning bolt, forcing the energy to the spark gap. A few hundred volts will still get through that. So following that there's a gas tube protector, which is essentially a neon tube which will ionize and short to ground. (Phone lines also have those at the central office end.) Following that is a MOV, as in a surge suppressor, to dump the remaining surge into ground. What's left after than can be tolerated by most RF electronics intended for such applications.
If this didn't work, radio wouldn't work in Florida. It's not that this stuff is expensive compared to the equipment it protects. It's that the front end stuff is big; #4 copper cables, big spark gap units, heavy ground rods, and solid metal equipment enclosures with welded seams.
Somewhere right now, a cellular tower is taking a lightning hit and restarting itself without damage.
You seem to know a lot about this, so all I can do is reiterate that when we take our electronics to florida to be tested, our QA guys make it clear that no company's electronics are designed to survive a lightning strike - including ours. (Though we are designed to resist a 22 calibre bullet). We install a lot of lightning arrestors, mandatory in Florida, São Paulo, Singapore -- and we seem to do fine in those areas (knock on wood) - but I've been told that's to prevent a nearby lightning strike from running down the antenna line into our electronics. If our RF gear (which is mounted outdoors in weather rated containers) gets hit with a direct strike - 100% guaranteed destruction.
I think you and GP are actually agreeing. The goal of a lightning protection system is to ensure that when a structure (antenna, wind turbine, transmission tower, etc.) is hit by lightning, the lightning energy is provided with a low impedance path to earth such that any sensitive equipment inside or attached to the structure is not exposed to currents and voltages beyond its design limits.
There are other considerations as well for personnel safety, like ensuring that the ensuing ground potential rise doesn't expose someone standing near the earthing point to a shock hazard.
Going back to the original comment, lightning protection in wind turbines is more or less a solved problem. Direct hits on blades are routine and any utility-class wind turbine will be designed with an appropriate system for the area in which it is located. Now, when your turbine is on a big rock, that's a problem and you may have to blast and backfill to achieve a sufficiently low earth resistance.
From all i have read, direct strikes on gear is a grab bag of destruction. I've read that in the Vietnam Conflict guys would get shot and the bullets would bounce off of their skin [strange effect of velocity, turbulence, and chance], so i bet there are examples of "look mom, no destruction" but i tend to side with you, at least as the guy managing this project: at no point am i going to say "cannot fail", "works everytime" or any version of the two. If you have never been to Maine, it is beautiful. If you have never built in Maine, don't get attached. The sea wind destroys everything. I kept my tools covered every day on a 42 day stint except 1, and the rust is ubiquitous. So i build to fail. The real question is, how long can i draw out the failure?
Like the other poster said, lightning can be dealt with, but I'm very glad I don't have your problem. You are dealing with the worst possible combinations of landscape and expense. My honest answer is you need a specialist with the proper testing equipment measure the resistance to earth at your location. If it is very high you will likely need deep holes dug (reading about one mountain top location that needed 600' to get proper grounding, you shouldn't need anything that extreme). The combination of a chemical grounding system in a well shaft may overcome your grounding issue. But, without that proper ground any lightning protection on your tower is apt not to be the shortest path to earth and will likely fail.
Judging from the (limited) SO responses, coupled with what i was told by the guy i am working for, the degradation of the cables makes the idea cost ineffective. Also, and i cannot stress this enough, in Maine you do not fluck with the Lobstermen. You don't even do anything that could be construed as tampering with their routines. Dudes are the hells angels of the small boat world. The thought of trying to explain why i look like i am cooking their profits before they can haul them up gives me the willies...
the water is a flat 40 F. average depth is 50' in the center. The Reach has an average flow (in/out) speed of 6 knots (underwater). The Reach moves about 1 trillion gallons of water every 6 hours between high and low (may sound hyperbolic; look up the Moosabec Reach). The tidal variation between high-to-low is ~14'. In a perfect world i would submerge a tidal generator in the deep point and power my island project AND Jonesport-Beals, but as i stated before Lobstermen do not like change, so i am pretty sure a giant underwater propeller is not on the menu... i digressed; the only swimmers are the seals, the sharks, and the daft.
So i am going to keep a journal on this project. I must be honest that the windmill overhaul is at the top of my punt list due to the multifaceted danger [climb up, remove topper, winch down, climb down, reverse&repeat, inline battery array, wicked wind in the west] but there are a number of other cool things i am doing there: rebuilding an articulating dock, re routing the water system, planning and implementing a solar system, repairing roads, rewiring structures, and preforming black magic rituals to curse the monsters at HughesNet [jk but serious]. We are going to play around with using Pi s or Arduinos for some of the management tasks, and the Ranger Power Management system is pretty cool, if dirt simple and dangerous to mess up. All this to say, if you are interested and have any advaice for my team and i, please hit me here. I will start posting a journal and pics when we start Phase I. The island is Norton Island in Downeast/Acadia right next to Jonesport-Beals [setting for SKs The Fog]. The organization that runs the writer's residency can be found here: http://www.easternfrontier.com/ aaaand i just realized it is EFF... lol. No relation.
We know where to put the waste, lots of the issues are political rather than technical. Lots of NIMBY with this stuff. Fusion also produces far less actual nuclear waste than fission if we can ever get that working. There are also working designs like the molten-salts that have failsafes that work without power and have no disastrous consequences (like radioactive fallout).
So cold fusion (as in cold enough to run on Earth, not cold enough to power your laptop) may compete with it, depending on what it will end up being. Unlike fission, fusion requires very specific parameters and if those are not present the reaction stops. My understanding and memory of this subject is rather dim, but I remember discussion benefits of fusion over fission as a potential power source in college. Relevant Wikipedia article: http://en.wikipedia.org/wiki/Fusion_power#Magnetic_confineme...
Cold fusion is totally the wrong term here, no matter how you twist it.
Artificial fusion actually requires higher temperatures than fusion in the core of the sun[1] because other parameters are less beneficial than at the center of the sun.
Actually; cheap storage makes the argument for nuclear energy even better. Nuclear is very bad at responding to demand because it can take weeks or months to restart a nuclear reactor.
Under any scenario, you're going to need some kind of power grid. That grid is going to have a non-zero "base" load as batteries charge, etc. Nuclear is perfect for generating this base load - it's reliable, clean and not subject to the variability that solar and wind are. Don't get me wrong, solar and wind are going to be a big part of the grid, but you want diversification of technology so that if you have a cloudy, but not very windy day, the grid can still keep up.
Also, fission reactors are being miniaturized, and at the timescale we're talking about, fusion may be an option as well. I don't think nuclear will be 50% of energy generation or anything, but it's a good, reliable technology that can supply a consistent amount of energy in any weather conditions.
The problem for nuclear is that with the way the bulk electricity market works, shaving off the peaks could hurt the ROI of building a new nuclear plant significantly. In New York and New England, suppliers bid in how much power they are willing to supply at a given price, the market operator runs a giant LP solver over the bids and transmission constraints between locations in order to satisfy demand in each location, every supplier that is scheduled in a given location gets paid the price of the marginal watt in that location. So nuclear plants bid negative to make sure they get scheduled and rely on other positive-price sources getting scheduled in order to turn a profit.
Those economics change drastically when cheap, mass storage of electricity is available. Spend all day charging your batteries with solar and wind, then discharge them in the evening. You could afford to underbid conventional fuels with renewables at similar margins and still negative-price nuclear. But that's the great thing about capitalism: it adapts to small efficiency gains in commodities production very quickly.
Storage shaves off price valleys, as well, by adding demand in off-peak hours. Every kilowatt-hour that's low enough under the peak price will convince someone to buy a battery to time-shift the power.
My model is that cheap batteries would drive on-demand power stations out, in favor of batteries + base load.
My understanding is that nuclear power is used to serve base load, which is the demand that isn't subject to significant variation. This allows generator owners to run nuclear generators at their cost-effective maximum output.
It's true that there is great variation in electricity pricing, with peak shaving and other so-called "ancillary services" bringing as much as 100X as high a price as base rate (for a short period of time). Nuke generators aren't capable of rapid load following, reactive power correction, voltage support, or other fast-response services that bring premium power prices.
The issue with base load contracts is that they are increasingly being undermined by power that is much cheaper, and reliable almost enough to substitute for base power.
Generally speaking most jurisdictions would have to be stupid to sign up to new long term (20 year+) base power contracts, but that's what nuclear power needs to pat for the high upfront costs.
As this article shows, new storage tech and load management tech makes it likely that base load will be less and less important over time.
Yeah, but the long-term outlook for fossil fuels is still rising prices. If given the choice between building a new LNG plant or a new nuclear plant, are LNG fuel prices low enough and stable enough over the long term to justify over a 20 year+ contract? This is an honest question and I'm sure the energy industry has a legion of forecasters and quant jocks on it, but I personally don't know how the economics work out.
There's also the carbon footprint aspect to consider. If the "carbon tax" is successful, nuclear could be a more cost-effective option than hydrocarbon-fired plants in areas without a more stable "natural" power source like hydroelectric or geothermal. The petrochemical industry likely has enough political pull to neuter any carbon tax law, but in theory that's how it should work.
That's my understanding as well. Nuclear Power plants are base load, and are contracted to do such, with long term guaranteed capacity. Peaker Plants take on load that exceeds the base, are paid on the margin, but aren't guaranteed any particular load.
There are places where solar is less than helpful -- up here in Scotland, we're north of Moscow: in midwinter we get as little as 4.5 hours of daylight every 24 hours (6 hours here in Edinburgh, in the south). Also, we tend to live in apartment buildings: many dwellings, only one roof, shared gardens.
(On the flip side, we're in one of the world's best areas for wind and tidal power.)
The point is, the far north (think Scandinavia, Russia, the UK) really needs non-solar. And the far-far north is a lousy environment to go out and fix a broken wind turbine. So there's probably a role for nuclear there.
There may also be a role for nuclear in shipping, although civil nuclear shipping peaked early (in the 1950s/60s) and the only folks currently doing much of it are the military and the Russian arctic icebreaker fleet. If oil becomes too expensive for propulsion, nuclear may be necessary as backup for wind power (sails and weather satellites work great together -- until you're becalmed).
Sweden in 2014 produced about 64 TWh (42%) from large hydro and 62 TWh (41%) from nuclear. The rest came from wind 12 TWh (7.9%) and from other 13 TWh (8.5%)! other being mostly biofuel and waste.
The interesting thing here is probably the change over time. In 2005 the distribution was 72 TWh (hydro), 70 TWh (nuclear), 0.9 TWh (wind), 12 TWh (other). With wind growing by 13x and the others staying relatively static, and energy use going down slightly. [1] There is quite a lot of variance per year, as the winter weather changes energy consumption quite significantly.
Electricity export has gone from about zero (2005-2007) to 15 TWh (2012-2014).
I don't think that's quite true. You don't need a lot of storage to make the peak-shaving, time-arbitrage setup he mostly talked about economic. But you need a whole lot of storage to make a reliable grid out of unpredictable generators. The author briefly touched on this in the context of home battery usage:
"Solar + a small battery may get someone in Germany to 70%, and someone in Southern California to 85%, but the amount of storage you need to deploy to increase that reliability goes up steeply as you approach 99.99%."
I believe the same logic applies to grid-scale storage. Getting rid of those reliable, dispatchable base-load nuclear and fossil-fuel power plants is going to be very expensive.
Base-load nuclear is the opposite of dispatchable. Dispatchable means you can turn it on and off at will. It takes a long time to shut down a nuclear plant and start it back up again (on the order of days not minutes). By contrast, natural gas plants can be highly dispatchable hence their use as operating reserve in many ISOs.
Hm. I was about to tell you that you're wrong, but then noticed you're an electrical engineer... so I guess you probably know more about this than I do.
But, my understanding is that dispatchable is the opposite of intermittent generation (solar, wind), because you can choose when it's running or not. Also, baseload is the opposite of peaking, because baseload generators take a relatively long time to turn on/off.
Thus nuclear would be dispatchable, baseload power since you can decide when it runs, but it takes a long time to get there.
Well yes, it's dispatchable in that sense. If you frame dispatchable as a binary property, it's certainly dispatchable in the way that wind is not.
But if you think of it from a power system planning perspective of economic dispatch (not just, can I dispatch - but can I afford it), nuclear isn't really there. Yes, you can E-stop a nuclear plant in seconds, but the cost is astronomical. Some nuclear plants like Bruce NGS in Ontario have thermal bypass - this increases their dispatchability by allowing them to dump steam and reduce electrical output quickly without touching the thermal output.
Nuclear's dispatchability doesn't really counter-act wind or solar's lack thereof since the time scale you're looking at is much different. Being able to start and stop my nuclear plant in 72 hours doesn't really help me if the wind stops blowing for a couple of hours. So in practice, you'd build other, more easily and economically dispatchable assets to meet your needs.
My company does work in the UK, and we're putting together communications system that will allow the wind generations systems to stop loading the network when the transmission lines get too hot (apparently it's less expensive to shut down the wind power x% of the time, than it is to build new transmission lines).
I'm wondering if you have any insight into what happens when our comms system sends a signal to those wind turbines and tells them to "Stop" - it has to happen fairly quickly, we have working factors of 15, 30, and 60 seconds, at which point we start escalating and instructing groups of wind turbines, and then eventually the whole farm to cease production. The wind turbines also have a keep-alive that has them auto-shutdown if they lose comms, as obviously it's far more important that the transmission line isn't damaged, than it is to stop producing energy for awhile. (From the perspective of the Distribution Utility, obviously the private turbine owners take slightly different perspective).
I see you have familiarity with wind farms, and I'm wondering if you know what the turbines do - do they free spin? Send load to ground? Come to a halt?
The primary way that a wind turbine "stops" is by pitching its blades (in some vendors' terminology they call this "pause" because the machine is still connected to the grid just not producing any energy). On modern machines this can reliably be achieved in less than 3 seconds. There is also a mechanical brake that can be applied for emergency stop in 1 second or less. However, mechanical braking is not to be used in normal operations.
However, going to the maximum ramp-down rate (which for a large wind park can approach hundreds of MW/s) is usually not desirable because it will impact grid voltage. At most wind parks voltage control is accomplished using on-load tap changes of the main transformers as well as switched capacitors. Both of these need a time delay of at least 10 seconds to avoid wearing out quickly so if you ramp your whole wind farm from max output to zero in 1 or 2 seconds, you will see a big impact on grid voltage. So if we need to ramp down for a transmission system thermal constraint, we will do it at the slowest possible rate that still meets the transmission utility's needs.
Ah, that makes perfect sense. It also explains why we have an expectation of a minimum of 15 seconds before we see any response to a command to stop loading the transmission lines, and why the emergency command to the entire wind-farm doesn't occur until all other options have been exhausted.
It makes sense to have the base load be something that is easily dispatchable and reliable. Renewables are an option in the short term, however the holy frail is nuclear fusion, which is clean, reliable and relatively cheap (building the infrastructure). In the short term, it actually only makes sense to replace generation that becomes retired with renewables because of the carbon costs associated with building new infrastructure. It makes sense that an inefficient natural gas plant that runs a couple of times a year at extreme peaks, will have a smaller carbon footprint than building new infrastructure.
Given gobs of cheap electricity, what are the options for synthesizing methane? Storing huge amounts of methane and using it to handle grid load is already a solved problem.
Synthesizing methane from biomass is a solved problem at a known cost that can't really be scaled away. It's not significant in the market because it's far more expensive than fossil natural gas. It's also more expensive than coal gasification, which is probably the second round in that particular gun.
So we could see peak load plants powered by gas sourced from filthy coal. More likely than a clean biomass solution, sadly.
>> We are basically drenched in free energy, but we can't store it cheaply enough. If that's fixed, we're done here.
I think that's a simplistic view. Yes there is tremendous energy in the movements of air in the atmosphere, and water in the seas, even more in the daily exposure to sunlight, and quite a bit in the geologic forces operating in the earth's interior. That's all well and good, but what matters in terms of powering a modern civilization is energy density, portability, and durability. Electric power isn't going to get a lot more portable than it is now. Storage improvements will make it more durable. But in terms of density the only sources we have are heat cycle sources dependent on fossil and nuclear fuels. Storage advances don't magically make renewable sources dense enough to replace the power we now get from non-renewable sources.
Edit: I should also have mentioned the obvious non-heat cycle source, hydro, which is dense enough in some places to be a very significant contributor.
This is entirely correct and always forgotten by the 'energy is everywhere, we just have to grab it' thinking.
The point is that wind energy is too diffuse and the capital costs of tying up that much land in wind generators is astronomical - and there is a very real degradation in site quality as you use up all the good sites first. It's pointless to say it's always windy somewhere, because transmission losses and costs make that unfeasible.
Solar is even worse because at even the best site on the best day you only get 50% of energy generation time, so event with perfect storage you have to have 2x the generation capability.
All this is the reason why - even right now with massive subsidies and compensation and special protection from environmental standards other developments have to adhere to - wind and solar account for about 1% of worldwide energy generation.
There is a limited role of grid top-up for solar and wind, but the future of energy generation is not either of them. The near term is in more efficient and less polluting modern gas and coal plants, with conventional nuclear filling increasing in use to provide clean energy without airborne pollution.
If the public can ever get over three decades of nuclear hysteria, that is.
I don't think thats is entirely correct. The issue isn't just storage but also production/output of the actual energy and as someone else once said here. Green energy is a linear solution to an exponential problem. I think there is something to that.
Also it doesn't solve anything for space travel which I don't believe are going to be put on hold.
> If we do get cheap scalable energy storage, it's pretty much game over (in the long run) for everything but solar and wind. We are basically drenched in free energy, but we can't store it cheaply enough. If that's fixed, we're done here.
You still need to get better return on capital before wind and solar are competitive on cost alone. Wind turbines don't last forever (and are hard to recycle well, since they need to be made of tough materials, which leads to some interesting but probably scalability-limited schemes to reuse old blades for things like bus shelters and playground equipment -- look up Wikado Playground in Copenhagen, for instance) and neither do solar panels.
Maybe that's also something they'll fix "in the long run" but it's really not a fundamentally different class of future-problem than improved storage.
Well one stumbling block is that we need to get thousands more people trained to service these energy storage facilities, let alone the windmills and solar panels. There is already a shortage there as the skill set is not one side, part electrician, part mechanical engineer, and so on.
With automation cutting more jobs in other industries, this could actually be beneficial. I don't think that training someone to be a serviceman on a solar plant takes as much pain as e.g. training a good programmer or a mechanical engineer.
There's no reason you couldn't bring up a infrastructure for technician level training to get those spots filled. I think you could just tap into the bureaucracy for existing job re-training efforts and it would work out fairly well.
If Tesla is building cells on a massive scale at the gigafactory, nothing stops them or a partner from building these into cargo-container utility scale storage systems and shipping them by rail to strategic utility interconnect points.
Sure. Cheap superconductors would also be a "black swan" for energy markets. They could enable trans-oceanic power lines for example, which would allow solar in Africa to power Manhattan at night. You could create a planetary power grid, which seems to me like something that would almost define a Kardashev type I civilization.
Fascinating. Never heard of this categorization of civilization technology before.
Apparently, according the wiki, in 2012 we were at type rating 0.724, presumably a large portion of that is non-renewable sources like coal/petroleum/natural gas.
Interesting that TypeII is almost a quantum leap over the TypeI - construction of a dyson sphere sounds like an almost unimaginably massive undertaking compared to blanketing the globe with solar cells.
Partly, but it isn't just storage, it's also the losses in transmitting the energy from A to B, especially if we're discussing areas which are relatively dim or windless.
I still need to be convinced that those two methods can produce enough to even reach a point of surplus to be stored given our energy needs and projected future needs.
If there's a legitimate arbitrage opportunity for individual homeowners to buy a battery and charge it overnight to cover peak costs, there should be an even stronger case for energy companies to buy larger batteries, install them in low-land-cost areas, and do the thing themselves. If these peak plants are so expensive, then a major utility can save money using batteries rather than peak plants.
The cost should be lower for a utility to do this than for individual homeowners to do so for equal capacities.
So I ask are we seeing this behavior?
On a slightly different note, it doesn't seem that the article is addressing vehicles. Battery powered cars are not yet practical enough for the mainstream, and aircraft are still entirely dependent on the energy density of hydrocarbon fuels.
I could see batteries becoming feasible for cars "soon" with the current rate of advance, but charging stations don't make sense to me. You want easily accessed battery packs in cars and battery-swap stations. This is a huge infrastructure change, but for out-of-city travel (i.e. a trip where you would need to charge before you get to your destination), you want the equivalent of a gas station, which is 1-10 minutes for a stop, not 45+.
As for aircraft, I don't see them using batteries soon.
Not quite "energy arbitrage", but ~hour long "frequency regulation" that was apparently once the domain of natural gas peaker plants are now being handled by a few flywheel energy storage designs.
"Energy Arbitrage" over a 24-hour period could be next.
Some home owners are doing this. Charge the batteries with solar in the day and top up with offpeak if more is needed for the early morning (say if you are using electricity for heating in winter). Not entirely off the grid but minimizing costs. I have talked to one industrious hacker who has two battery installations. He charges one off offpeak and feeds it back in for a profit in peak times while charging the other on solar and doing the same thing (because providers cap the amount of current you can feed back into the system as an individual). I am looking at getting myself a solar + batteries + grid system so it doesn't seem so far fetched that investors may be planning something similar.
> but for out-of-city travel (i.e. a trip where you would
need to charge before you get to your destination), you want the equivalent of a gas station, which is 1-10 minutes for a stop, not 45+.
I appreciate that 'time is money' and people want to get where they are going sooner rather than later... but I wonder if there are social benefits from a transport infrastructure that requires us to stop for 45 minutes to recharge, not just the car but the body? Could potentially result in fewer road fatalities from fatigue if technology enforces a break on drivers? And potential economic benefits for small towns to offer charging stations right by a place to buy a meal...
> but I wonder if there are social
> benefits from a transport infrastructure
> that requires us to stop for 45 minutes
> to recharge, not just the car but the
> body?
Would they be able to outweigh the social benefits (and reduced energy use) of traveling 45 minutes slower?
I think you missed the part about studies that make the case that utilities could be doing this today. Utilities are conservative though, I don't expect them to jump in with both feet right away.
> This is, of course, speculative. We don’t know if the study findings scale to the whole of the United States. It’s back of the envelope math. Atop that, the study itself is an analysis, which is not the same value as experience.
Hence my question. Do we actually see any utilities making steps in this direction?
They're openly talking about it. Let's look again in ten years, given the lead time, conservatism, and regulatory concerns at this scale. "Fail fast" isn't really an option for the grid the way it is for software startups.
I realize I may have come across somewhat pugnaciously above. I am honestly curious about this, though.
What utilities/where are talking about this right now? What level of talk are we talking about? Public statements? Designs? Construction permits in the works?
My research has shown that while many utilities have purchased storage (SCE 250 mW in 2014[1]) it's still very experimental and they do it mostly for research purposes or publicity or because the regulator makes them. However, since the electricity business is slow moving, low margin and very capital intensive it's clear why everyone is very interested in any new potentially disruptive technology. Imagine owning a peaker power plant that you've paid for in full using debt and generates a little profit each year under the condition that it can be amortized for 15 more years - if someone built the equivalent of a peaker plant made of batteries, even in 2025, you'd still go under. Scary.
In many states, utilities do not generate power they merely buy it. They call it energy deregulation.
If it really were possible, you'd see companies like Exelon attaching battery storage to their nuke and coal plants to feed it back at a profit during peak rate periods.
My guess is that isn't cheaper than using peaker plants yet.
One problem that will surface is that the cost of our transmission infrastructure will have to be covered, one way or the other. Currently, with net metering, customers are able to arbitrage power at the retail price, while effectively making use of the grid for both pushing and pulling load. This would be like flying a round-trip from San Jose to Tokyo and back to SFO, and claiming that you only owe the airline for the Caltrain fare up the peninsula.
The other issue is that our current infrastructure isn't designed for peer-to-peer transmission, with neighborhoods pumping large amounts of power into the transmission lines during peak solar hours. Some areas of Hawaii have had to put a moratorium on installing rooftop solar to prevent potential damage to the branch circuits.
Long story short, it's not simply a matter of storage costs undercutting retail power costs, there are also maintenance costs for the grid that are invariant on demand, which will have to get paid one way or the other.
I think you misunderstood something in this article. The only stored power that would be on the grid would be from the power companies themselves, not the end users. End users wouldn't resell the power, they'd use it themselves.
The growth of distributed solar + storage (i.e., zillions of residences having 10kW arrays and 25kWhr storage) is accelerating. Projections I've seen suggest that this distributed storage (which doesn't belong to the power companies) will be far larger than utility-scale storage such as discussed in the article. This presents real challenges in how to control that distributed storage and pay its owners for storage-related grid services.
Why can't it just be a feed-in tariff? And if the consumer can consume using the network they can produce, in the other direction. The cable runs both ways, no?
It can be, but that doesn't capture the value a consumer with fast response, controllable generation/storage supplies to the grid. Compensating the consumer for power or storage they supply at a particular time is complicated.
Existing net metering pays a fixed rate for power supplied to the grid by, say, a solar-equipped residential grid customer.
But in fact, fast response power delivered to the grid when the grid managers call for it commands much higher prices in the dynamic electric power market. So does the ability to store power on demand (absorb power from the grid), when generation briefly exceeds supply.
So imagine a residential grid customer with local solar + storage that's controllable by the ISO (grid managers). Rather than net metering (a fixed feed-in tariff), the ISO should pay instantaneous market rate for power delivered, or absorbed by, the residential grid customer.
I think we'll soon converge on bidirectional power delivery/sink services that are controlled by either the ISO or perhaps by local smarts on the resident's grid-tie interface. That would engage the resident's system as a grid stability enhancement tool, and would bring more revenue to the resident than simple feed-in tariff schedules. One assumes the resident would participate by an opt-in choice in exchange for enhanced revenue.
Thanks, very informative. The biggest challenge seem to be that the current people running both te grid and power companies are so conservative that they seem to be the biggest blocker against something line this being created.
IMO It only seems that way. They do their best to stay on top of all new technology, but as consumers are used to very high level of service (>99.99%) they can afford few risks if any.
All the more reason for local storage. The closer to the end user that the power is stored, the less load on the transmission lines. If you store it at your own house, it never needs to hit the grid at all.
Yes, that's a good thing. However, imagine a world in which everyone has local storage and only needs power on the 10 days of the year when it's dark and rainy. So, let's say the total power draw per household drops to 100 kW per year (about $15 at California power rates), vs perhaps 10,000 kW per year today (about $1500). Meanwhile, the cost for maintaining the infrastructure changes very little -- crews still need to come out and trim trees away from the lines, pull fried squirrels off of transformers, re-string lines after a storm, etc. Assuming half of your current bill is spent on infrastructure, the utilities would need to recoup $750 or so per year from their customers to provide that $15 in power.
It stands out to me that pumped storage only gets a passing mention. We've had that as a proven energy storage technology for decades. People in the industry love to talk about it and file FERC applications for preliminary permits on the same sites over and over (I would know, I've had to read nearly all of them over the last year at work), but it's never deployed at the scale people expected.
It'll be interesting to see if if PSH ever really takes off, or if it really does get left in the dust by batteries and other things.
Pumped hydro works very well. Its main issue is that the enabling favorable combination of terrain (adjacent lakes or potential lakes at differing heights) and water supply aren't common, so cost-effective sites are restricted.
It is most of deployed storage atm, but the bulk of that wasn't deployed recently. I'm staring at a very informative figure I just made last week that I can't share since it's a draft for a report, but if you glance at EIA 860 data you'll see the breakdown for storage by nameplate capacity is:
Battery - 0.7%
Compressed Air - 0.5%
Concentrated Solar Power Storage - 1.3% (which doesn't always exactly count)
Flywheels - 0.2%
Pumped Storage - 97.4%
You can also see that the average operational year for pumped storage is 1974, not anytime recent. You could weight the years by capacity, but most plants are pretty big, and it shouldn't change it too much. Also, pumped storage capacity only totals 21.6GW.
As for where I've been, I've just been stupidly busy. I keep meaning to jump back in, but I never seem to have the time.
>This leads to what seems to be a paradoxical situation. A battery that is more expensive than the average price of grid electricity can nonetheless arbitrage the grid and save one money. That’s math.
Missing in the math is logic. If enough people do this, the price will even out and the arbitrage opportunity will vanish.
It's like an arbitrage trading a model that makes money when paper trading but loses money when live because the effects of live trades on the market wasn't considered.
The fact that an arbitrage opportunity will not always be available does not mean that arbitrage opportunity does not exist.
Also electricity companies would be very slow to adjust to reduced demand by lowering prices, making this work for longer than a naive analysis would predict.
>The fact that an arbitrage opportunity will not always be available does not mean that arbitrage opportunity does not exist.
Spotting a leftover cookie on the table means you will probably be able to get a free cookie. It doesn't mean that you have found a way to solve world hunger.
Everything you have said can be said about literally every possible opportunity to make money outside of those that leverage sweat equity.
If you want to say something about his idea that is specific you could add value.
For instance attempting this arbitrage is a risk transference since you are betting that the technology of batteries won't catch up before you have made back your capital investment.
The arbitrage opportunity won't vanish, but rather it will diminish until the cost of performing the arbitrage roughly equals the money you can make from it. When the peak/off-peak price differential drops such that buying more storage isn't profitable, people will stop doing it, and the price will stabilize. It's just like any other normal market: prices stabilize at something near the costs.
Also, it will usually be more profitable to build more generating capacity, reduce demand or improve the grid. You can build massive arrays of expensive batteries but you probably don't want to.
The other reason energy storage could get big in the next several years is an elevated risk of blackouts due to aggressive retirement of coal-fired power plants -- blackouts like the one in Washington last week that shut down several government agencies (e.g. the Department of Energy) and which was probably related to the shutdown of the Potomac River Generating Station.
Nuclear plants take years to build and have shitty failure scenarios. Best to stopgap coal until renewables take over with combined cycle natural gas plants, fueled by cheap natgas from fracking.
It's not perfect, but those natural gas plants will still be kept around as peaking plants and they're an order of magnitude cleaner than coal plants (as well as producing much less CO2 per unit of power generated).
> Nuclear plants take years to build and have shitty failure scenarios
That's only because we won't build any newer designs that don't have shitty failure modes and insist that we keep old reactors with very shitty failure modes online, though.
I don't disagree that we can build better designs. I disagree that we can build them quickly and cheap.
According to the OECD:
"As nuclear power plants are complex construction projects, their construction periods are longer than other large power plants. It is typically expected to take 5 to 7 years to build a large nuclear unit (not including the time required for planning and licensing)."
"Between 2002 and 2008, for example, cost estimates for new nuclear plant construction rose from between $2 billion and $4 billion per unit to $9 billion per unit, according to a 2009 UCS report, while experience with new construction in Europe has seen costs continue to soar."
All the modern design proposals are to mass-produce smaller units -- sometimes, completely sealed units -- which would then be installed several to a site, and maybe sent back to the manufacturer to be refurbished at periodic intervals. This is the "save money by mass production" economies-of-scale approach, as opposed to "save money by being more efficient with large custom-built plants" economies-of-scale approach which has failed the industry and led to many of those ballooning costs.
(Whether we have the regulatory wherewithal to go with that plan is another matter, but the very idea of reusing the same design over and over does help keep compliance costs down, and you can focus on site-selection costs.)
How does that compare with other sources? 5 Years and a few billion doesn't seem that bad to me. Even plans for building a few thousand new homes can be over the next ten to fifteen years.
Surely the main cause of failure is human rather than the actual design. Nuclear stations may be safe, but that comes at an increasing cost in money and complexity. How do we deal with the institutional failings that allowed us to build "old reactors with very shitty failure modes". Stop blaming it on the tech. I want reactors that are simple enough to be cheap and easy.
Humans will always make mistakes, even within the design process. Nuclear is a dangerous technology to make a mistake with.
Diablo Canyon Power Plant was partially built backwards, by mistake. It's absurd.
The company updated its plans and added structural supports designed to reinforce stability in case of earthquake. In September 1981, PG&E discovered that a single set of blueprints was used for these structural supports; workers were supposed to have reversed the plans when switching to the second reactor, but did not. According to Charles Perrow, the result of the error was that "many parts were needlessly reinforced, while others, which should have been strengthened, were left untouched." Nonetheless, on March 19, 1982 the Nuclear Regulatory Commission decided not to review its 1978 decision approving the plant's safety, despite these and other design errors.
The plant had peak-load units which would have been spun up, if they still existed. (Alternatively, if the grid had lots of energy-storage devices around, it would have used those instead.)
One thing that kept coming to mind while reading all of this article's details on different methods for storing and selling and buying power at the peaks and valley of pricing was this:
Electricity in our homes is so widespread and popular because it just works. We don't need to think short-term about how or when we use it (although we should and can if you want to). You flip a switch, the light comes on, and you flip the switch again when you're done.
In my mind, there are only two factors that need to be there for widespread adoption: the price needs to come down, and the battery or other storage medium needs to just disappear into the background of in-home electricity usage. Most people will just want to see a lower bill without any costs to their ease of use regarding electricity.
Every time I hear about advances in renewable energy I get a little sad. I've rented for the last decade plus, and unless I get married and decide to settle down, will probably continue to rent for the next decade plus.
Homeownership would allow me to take advantage of these sorts of advances, but so would the ability to take advantage of this as a renter.
Where I live, in Sweden, you can. I purchased shares in a wind power coop. It feeds me all the electricity I need wherever I live in the country, at cost of production. And with interest rates at effective zero here the investment is significantly better than having the money in the bank.
A quick and dirty calculation: I bought 28 shares from OX2 windpower coop, which entitles me to 28,000 kWh at cost price. Which is what our house/home office uses per year. The price is about €700/share. But I bought shares from the market (people that want to sell their shares), which for some reason are cheaper. I paid about €590/share.
Over the last seven years (I haven't had my shares that long), the saving on cost of electricity, which is tax free, would have been about €960/year. Which is about 5.7% ROI /year. Better than bank rates but worse than index linked stock market investments (I think).
I don't think OX2 offer electricity in Germany, but maybe in Finland and Poland. Even if they did, it would also depend on how the German regulator has structured how this can work, as well as how taxation on this is structured.
Your landlord, however, has an incentive to install solar (and claim the relevant tax credit) in order to increase the attractiveness of his rental by advertising lower utility bills.
Except it takes a really big train to weigh as much as a lake.
Pumped hydro is unbelievable. Tour a pumped hydro facility if you ever get the chance, and ask about the economics. They're buying a dollar of energy at night in Spring and selling it on hot summer days for something like five hundred. (Probably not quite that much, but it was a crazy ratio, to the point where efficiency barely matters - also, free energy from rain!)
The ARES website claims that the cost of their train technology is only 60% of an equivalent pumped hydro facility [1]. This is pretty impressive, and I wonder how they do that. Maybe because a water pump is not just a turbine operating in reverse, and thus pumped hydro needs separate equipment for charging/discharging, whereas ARES can use the same?
From site: "3GW Regional Energy Storage Hubs. Energy capacity may range from 4 to 16 hours duration at full power output."
The 3GW is big but the duration is small, so I'd guess they're doing short term grid stabilization, competing more with flywheels than hydro. That's just a hunch, but if that were true, it'd explain how you could get something really effective without a lake-sized train. (Some videos online suggest they move a bunch of heavy cars on the same track independently, so they probably also get much more weight than a normal train.)
My limited understanding of pumped hydro specifically was that it really was basically just a turbine operating in two directions, crazy as that sounds. Even so though, I can easily imagine big cost savings just from not having to engineer around water. That stuff doesn't compress, you design or run these wrong and they'll just mercilessly explode:
https://www.youtube.com/watch?v=6hVUeNp3o3M
Being able to build anywhere is really neat too.
So, grid storage tends to fracture into a half dozen specific types of storage, all based on duration, spin up time, geography, fuel, etc. There are like 18 different tradeoffs, and you can specialize in any one of them. Pretty much any new grid storage technology is going to add to the field rather than displacing anything, because the grid needs arbitrage in basically all 18 different dimensions.
Thanks for the link. There are some cool ARES videos on youtube too.
It actually can be that price multiple. Pumped hydro is one of those fast response storage technologies that can supply power for high price premium ancillary services (transient supply/load matching, peak shaving, etc.).
> Most flow battery companies have $100 / kwh capital cost as a target
With typical power plants generating e.g. 500 MW 24 hours long, the storage is going to be 100000 $/MWh * 500 MW = 50M/hour. If we assume that a solar plant generates power for 12 hours, and the batteries feed the consumers for 12 more hours, it's going to be 600M for storage alone. The solar power plant with a peak power well above 500MW (to feed the day load + charge the batteries for the night) will cost you extra.
I still think that a Thorium molten salt reactor is a strong contender in a price landscape like this. (It also has a nice ability to burn our current stockpiles of radioactive waste from Uranium reactors.)
One thing to add: Electric cars. They come with energy storage in-built.
By rough calculation, if 10% of cars in my city (Perth, Australia) were electric, their batteries could supply the entire city's demand for duration of about 2 hours (or 10% of demand for 20 hours etc). This could work really well for demand balancing and peak shaving - overcapacity (which in Perth is massive, since the policy is to maintain supply even on extremely hot days, when demand shoots up and generation capacity goes down) and spinning reserve could be tremendously reduced.
I suspect some good software and a little hardware will be needed to account for the owners' needs optimally.
Silly question perhaps… If alternative energy sources in combination with batteries increase independence from those who supply it, and reduced demand during peak times allows rates to fall, isn't there a point where investments in infrastructure (or maintenance thereof) will stop being profitable, etc etc.
Yes. This is the grid owners' nightmare at the moment. Their old business model is rapidly dissolving as the price of solar + storage falls ever further and more electricity users essentially disconnect from buying grid power. That revenue loss hits grid maintenance expenditures directly.
Random idea, if batteries are cheap enough could we charge them up and ship them on trains back forth instead of building high voltage transmission lines?
I had the same thought reading the article, but after some thought, I doubt it.
Grid transmission can be as low as 0.5% per 100 miles already. Batteries lose about 10% of the energy by storing it, that leaves you with needing to go at least 2000 miles in order to make up the battery loss difference, plus needing the energy to move the batteries physically.
Batteries are cheap to charge in-place. Once you load those into a train you have to add the cost (both economic and in energy conversion rates) of moving the whole mass around in a diesel powered vehicle.
We already do that with alkaline batteries, and we pay premium price per BTU because of the convenience of begin able to carry small quantities of electricity around. But if you want to move city-powering amounts of it (think millions of Megawatt-hour), I am guessing a grid is more efficient.
Probably not since most radial power systems are not designed to support islanded operation for a segment of the system. Grid batteries will likely be designed to automatically disconnect during grid outages just like distributed generation.
How lucky for me, I've just finished doing the same back of the envelope calculation for using a home storage system that would take advantage of Time-Of-Use rates, my conclusion was opposite - when he is concentrating on the gap between ~34c peak rate to ~15c minimum rate I've chosen to focus on the gap between the minimum rate to the ~19c standard tiered rate. the 3-4 cent difference means that you can only save, at the maximum, 3-4 cents per kwh consumed your storage. And storage is very far from this range.
Nonetheless, I'm still very optimistic about other uses of storage and I am confident that I would find the right company to short, maybe a company that builds peaker plants.
Shocked at the amount of support for nuclear here. Use the data - how much solar, wind and nuclear capacity had been installed in the last 5 years?
Nuclear is not getting cheaper, not significantly, so what does this imply for the next five years? The answers are obvious, and non nuclear.
New Zealand, where I live, declared itself nuclear free in 1984. We are currently generating 70-80% of our electricity using renewable/low/zero emission methods.
Apologies for the mobile link, but [0] says that in 2014 over 50% of your energy was from oil and coal. Have you a source on your claims? I know that many of the non-nuclear countries in Europe end up importing a lot of their power from neighbouring countries like France, who's main supply is nuclear.
The big advantage of nuclear is it's relatively safe and readily available. How many people have been killed in nuclear related incidents in the past 30 years vs people killed in coal mines, for instance?
If you haven't read Ramez's books, you're missing out. Check out Nexus, and then the sequels. Without a doubt, the best sci-fi I've read in a long time
Interesting though the Compressed Air storage could have some "interesting" failure modes.
Back when I was an Engineering Technician my thermo fluids lecturer commented that if the main reservoir for the labs compressed air system burst it would flatten the entire lab block.
Got any real numbers to back that up? I'm guessing that it just feels good to bang on your chest and make that call. The reality is probably a lot different. How are all the megacities going to get enough solar and wind power, for example? Tokyo, NYC, Mexico City, Shanghai, etc. all running from solar? How big will the grid need to be to power NYC? Anyway, I'd love to see a deeper analysis and less chest pounding.
That part of your comment wasn't off-topic, but the part about "chest-beating" was, and it provoked a wildly off-topic digression. Had you followed the HN guidelines and limited your comment to its substantive and civil portion, I don't believe that would have happened.
One little jab in the entire thread by me. Not mean spirited but an accurate description of what people do.
All in all, I thought I was quite patient and I explained why I thought we can, and should try, to have more intelligent discussions on HN. I'm afraid that I didn't convince the other guy.
"HN comments sections cannot be more than superficial"
Anyway, I'm sure the moderators will figure it out before we become TIL on Reddit:
I take the point, but one thing we've learned in trying to have a sizeable online forum where the values are 'civil and substantive' is how destructive the little jabs can be. If we're to have the intelligent discussion you speak of, we all need to edit our jabs. It takes discipline, and there's a cost (a certain blandness), but it's necessary. I've had to learn this the hard way, and am far from the only one.
I've often wondered why jabs, swipes and sarcasm are so corrosive on HN when anyone who knows about the history of discourse knows what a lively role they have played. Literary wit, especially, is frequently scathing—both an art and a sport. Not having an answer to that is why it took me so long to learn not to engage that way on HN.
But I think I know the answer now. It's because this community is many orders of magnitude larger than those were, and so the lowest common denominator is both much nearer and much suckier. Keeping that at bay is HN's major task, and the main purpose of the site guidelines. It makes the discourse a little more bland, but the alternative is not lively exchanges of high wit, it's YouTube comments.
The point regarding literary wit is well taken, and I think it works so wonderfully in "real" literature is the difficulty and delay in responses. Folks have to be pretty vindictive to carry their gut angry response all the way through to publishing, whereas here, all I have to do is hit "reply".
HN isn't where you go to learn, HN is where you go to expand the realm of possible information for you to consume.
When you read a comment like the above, your response shouldn't be to demand for sources, your response should be to go research the topic on your own.
This site would be much better if folks understood that HN itself isn't a place for someone to "get wonky".
On my debate team, we had a rule of "no wonking". HN would be a better place if it too had the same rule.
It certainly could and should be a place to learn. How do chest pounding statements add value, especially when they fall apart after the first question?
There are definitely a lot of knowledgable people who read this site. Why wouldn't you want them to contribute and take deeper dives? Superficial conversations are essentially noise.
If you don't see a qualitative potential difference between a quick reply (HN's comments section), and a well thought out article (HN submissions), then I can't help you.
HN comments sections cannot be more than superficial. Anything more than superficial deserves to be put in a more accessible and readable place, as a topic in its own right.
Of course replies can be more than superficial. They can contain facts, historical perspective, personal experience, back of the envelope calculations, or qualitative statements from people with specific domain knowledge.
After a good HN discussion, there would be enough information for another blog.
No, they can't, because the folks who are commenting at what you think is that level are actually completely unqualified to do so.
The folks who are qualified to write articles and papers on the topics that might interest you, do exactly that. There is no inherent value in wasting intellectual effort on an HN comment. The people who comment on HN are not the people who are the most qualified to speak on a topic in depth, or if they are, their efforts within HN's comment framework will be very limited, and most certainly hindered by the medium, not to mention the audience.
I am personally aware of a handful of genuine experts on HN who comment on articles in their "wheelhouse", and they've never risen to the level of their capabilities while doing so. I don't begrudge them this, it's unreasonable to think they're going to put in the level of effort in an HN comment they put into their professional work, but to suggest HN comments could ever rise to the level of what experts are capable of producing is not only naive, but also greedy, and perhaps more than a little lazy.
If you want to know more about a topic, that's your responsibility. There are a number of ways around which one could wander to learn more about what's currently known regarding a topic, and the folks who are experts and are commenting in HN don't have the time to go through the proper motions of releasing new information. Even if they had the time/inclination, they wouldn't do it here.
If you take nothing else away from my comment, consider the following: This (HN comments section) is a back channel, if anything. You don't scrutinize a back channel, you verify it independently.
> No, they can't, because the folks who are commenting at what you think is that level are actually completely unqualified to do so.
> The folks who are qualified to write articles and papers on the topics that might interest you, do exactly that.
How could you possibly know who is qualified and who is not? Given that the authors of submitted articles frequently comment on HN themselves, it must be true that at least some HN comments are made by qualified people. But that assumes that writing an article is the sole requirement for being "qualified". Not all articles are written by qualified people. Additionally, many articles are watered down by journalists, and HN threads can provide broader perspectives.
> There is no inherent value in wasting intellectual effort on an HN comment.
There is as much inherent value as there is in wasting intellectual effort in a conversation at a dinner party. If I am talking to someone whose life experience is completely different from mine, I am learning something even though they are not rising to the level of their capabilities. I am allowed to learn subjects superficially out of curiosity and without the blessing of an authority figure. I have to choose who to listen to and who to ignore just like I do with other sources of information such as HN submissions.
Asking for sources in response to an unfounded claim is common on HN. I've never seen anyone called naive, greedy, and lazy for it, though. At any rate, asking for a source does not preclude one from researching the topic separately.
I agree that HN should not be put on a pedestal. Neither should the submissions.
I'm not a part of this conversation but I was assuming you were playing the role of the condescending nerd who tells other people how to think. was I wrong?
No, but you seemed to imply that one can only learn from certain people in certain contexts. I find that I gain surprising insight from unexpected people at unexpected times. HN is not an exception.
I wasn't implying anything, I was directly stating that the folks who are most beneficial to learn from aren't making elaborate posts on HN, and you shouldn't demand that they do.
You're just being lazy by demanding someone to do your research or write an article for you. Think of it from their perspective: why would they? Any reason someone might post on HN with a comment containing original or well-thought content can be better accomplished by putting that effort elsewhere, in a blog post or website article, and then posting that article as a submission to HN.
Only a small fraction of HN users even visit the comments page, not to mention the additional exposure one might achieve by getting their message out to folks who don't go to HN.
So yes, there are things to be learned from HN comments. Just not the way you're suggesting.
In other words, stop asking folks for citations on HN (or any other comments section, for that matter). You're wasting everyone's time.
I believe you should reread the previous parts of the thread, including your own, because it seems like your understanding of the positions in this argument have shifted. melling isn't demanding that arbitrary experts make elaborate posts on HN for his or her education and betterment. Rather, melling seems to believe that people who present information as knowledge in the comments section should be able and willing to explain the source of their certainty.
You're positing that people with expert knowledge have an inherent interest in exposing that knowledge to as large an audience as possible, and that this interest is so significant that they would always preferentially invest the extra effort and resources required to write a blog post or website article that properly contextualized and presented the information.
You're positing that the prospect of writing an article or blog post would preclude the possibility of responding to the comment on HN.
You're positing that the nature of this interest implies that these experts would wish to submit their own writings to HN.
To me, it seems that these terms must imply that this interest is not so much in the distribution of information, but in publicity and the tacit validation of expertise brought by the ownership of the information as an original source.
In my experience, this does not reflect the general behaviour of domain experts, who routinely correspond on mailing lists, usenet, and comment forums, including HN. There are in fact extensive comments made by at least one electrical engineer in response to this HN submission.
There is a subset of experts who already blog, for whatever reason, and stand to gain in some way by presenting themselves as an authortative source of knowledge and thought to the HN audience. However, many such persons (patio11 comes to mind) routinely write extensive responses to comments on HN.
Personally, the specific reason I began to routinely read the comments on HN was precisely because I could read fascinating domain-expert knowledge, presented in dense paragraphs, regarding topics that interested me. I usually read the comments about a submission before reading it.
Personally, I do wish to know whether someone's thoughts on a topic reflect domain expertise, or hobby interest; I do wish to know whether an assertion is established fact within a domain, controversial, sourced from an article or a blog post, et cetera, especially if that assertion seems dubious to me; the alternative is to simply continue believing what I believe, and to disregard such statements. In my experience, however, it is sometimes I who is wrong, and the dubious assertion which is correct.
In closing, I'd like to mention that melling, as the new OP of this now detached comment thread, was censured for describing someone's assertions as 'chest beating', which I think was appropriate. However, in the course of your comment here, you have accidentally or deliberately asserted that he was both "lazy" and "wasting everyone's time." I don't think that this sort of argumentation is useful, and I doubt that it is acceptable in your team debates. dang[a moderator]'s comment in reply to melling about "jabs" (https://news.ycombinator.com/item?id=9378899) is relevant, well-written, and interesting, and I think that you could benefit from reading it.
You've taken what I wrote and pushed it much further than I did. If you re-read what I've written, I'm not nearly as extreme as you present.
Your argument hinges on me being extreme in my position, when in actuality I am not. I have not said anything nearly as absolute as you've implied. For example, I don't believe me using the words "lazy" or "wasting everyone's time" were an example of what dang refers to as "jabs" -- I was describing an activity. My words were not used as personal insults, but as descriptions of behavior. It is lazy to ask for sources on HN comments, and it is a waste of everyone's time, for the reasons I specified. These aren't insults, because they're specific, and directed at an activity, rather than a person, and I provide reasons for why those activities are as described.
I'm sure some SMEs do make the mistake of posting on HN in an elaborate and detailed fashion, as another example. I've never said it's impossible, that'd be an absurd thing to say. I just said it's not in a) their best interest, and b) many won't for the reasons I outlined.
Coming to HN for the reasons you do is an absolute and unambiguous mistake. What you get here is bad quality, for what you're looking for. You can get much better elsewhere, in the form of blog posts, articles, wiki pages, published papers, etc.
I'll repeat my thesis -- HN holds value as a "back room" where experts can speak without the same rigors with which they usually have to speak. There is value in informal discussion, and that's what happens on HN. Citing sources is a waste of time on HN, because you shouldn't take anything you read here at face value.
Besides, next time you're in a conversation, are you really going to say, "a post I read on HN said..." as a way of citing the source of your knowledge?
Actually, you used the personal pronoun 'you', which is not used in English to describe an activity, but rather a person. This is unambiguous.
To me, your response on this matter indicates that I'm likely to get argumentation in reply to anything I say rather than discussion or honest reflection.
This is fine, if one likes argumentation for its own sake, which it is established that you yourself do; I, however, do not, so I'm going to disengage from this conversation.
Scanning your points briefly, I would in another context be happy to go back and re-examine my representation of your position, but will not do so, as I can't take the assertion at face value, and would anyways be wasting my time as per above. I will remember it as a possibility. I do not exactly remember your thesis in the terms in which you have now restated it, but it seems likely that we are largely talking at cross-purposes with regards to the nature of valuable discussion on HN. Experts seldom fail to speak with sufficient rigor or mind correction when they do so. And 'citing sources' can be a waste of time, or a courtesy to others, depending on the context.
I would not hesitate to link to information on HN if that was the source of my knowledge. It is more important to me that I be (eventually) correct than that I maintain an appearance of correctness.
This gets to another point that bugs me in these discussions... the pushback against wind/solar dominance by the pro-nuclear crowd. Insisting that nuclear power is the right way to go is western-centric. It's fine for the US, Europe, Japan, and other advanced nations that have the infrastructure to support it. But is it a solution for Peru? For Somalia? Of course not. Wind/solar/storage, on the other hand, is totally viable as a solution for even the poorest nations. This alone is an argument for solar over nuclear.