Hacker News new | past | comments | ask | show | jobs | submit login
Solar Surpasses Gas and Wind as Biggest Source of New U.S. Power (bloomberg.com)
362 points by adventured on June 12, 2018 | hide | past | web | favorite | 264 comments

Even with a tariff on panels, the cost of a solar installation, with incentives, is still significantly cheaper than grid electricity.

With federal and state incentives, my panels work out to about 4-5 cents a kilowatt hour. Normally I'd pay about 17 cents a kilowatt hour.

(This makes running a space heater cost about as much as running a natural gas furnace.)

Furthermore, my panels are financed with a fixed-rate loan instead of tied to market price of energy. That means the cost will never go up, but instead go to zero at the end of the loan.

Are you factoring in the price of your financing?

I very often see that ignored. You either need to include financing costs, or opportunity costs. Spending e.g. $20k isn't free, either you don't have it and need to borrow, or you do and you need to forgo putting it in the S&P500 and getting 7% on it long-term inflation adjusted, which is about $120 a month that you could spend on electricity. For comparison, the average US home uses 900 kWh per month, which is about $90 at average electricity prices in the US.

You can argue the numbers (e.g. 'you're wrong, the S&P500 has lower or higher returns' or 'I paid just 10k, not 20k of my own money'), I'd agree, but you need to consider everything. I find most calculations are often way too generous and omit important aspects.

That's not a reason to not do it, btw... I'd say, if you're underperforming the S&P500 by a little bit while going 100% green on your consumption, that's terrific. But you'd have to admit it's more than just a financial story.

The other side of the story is that you likely aren't consuming your own energy. i.e., you're not paying for storage or expensive peaker-plant backup capacity. If you were the only consumer, or if all consumers were like you, your panels alone would not be sufficient. In other words, there's a cost that has to be paid, and someone else is paying it now. At some point utilities are going to pay you 4-5c per kWh that you put back into the system, and charge you 17c. There is no reason to believe today's feed in tariffs will continue indefinitely. So whatever calculations you make today may not fly in the future, unless you buy your own storage or excess capacity, but then you're not going to get anywhere near the 4-5c figure.

I'm a big believer in solar and I'd love for us to commit more resources, even if it increases prices, it's too important. But we also tend to overstate how cheap everything has become, the narrative on solar is too positive, often portraying solar as if it's ideologically-neutral and financially-interesting when reality iis the opposite. We're still only at 1% of total energy production and we're subsidising it up the wazoo from state and federal incentives to utilities paying retail prices for your energy, and it's still not particularly interesting once you figure in opportunity costs. We still have a long way to go.

> Even with a tariff on panels, the cost of a solar installation, with incentives, is still significantly cheaper than grid electricity.

That all depends on where you live. I could cover my house with panels and not have enough in the winter. It would work great in the summer.

It also depends on how your utility bills. During the spring and fall where my bill is the lowest (low usage of heat or a/c) half of my bill is just for being connected to the grid. Since I can't ditch the grid because I need it in the winter, solar doesn't come close to paying for itself in my situation. Power is under 7 cents/kwh, but there is a nearly $40 a month fee for being hooked up to the grid.

That said, it is still exciting to see the price come down. I'll likely get a few panels just so I have a little juice when the grid is down.

> 'll likely get a few panels just so I have a little juice when the grid is down.

Most grid tie inverters have a shut-off switch that kills all power in case of a power outage. This is usually mandated by safety regulations, to protect linemen working on the downed lines. (there are a few that provide very limited solar power during an outage, but not full power, something like 2000W)

Yes, a point to keep in mind when buying equipment. I had a very disappointed neighbor when I finally convinced him to shut the power off to his whole house and see what would happen. The salesman sold him on having power when the grid was down.

It takes more work, planning and some battery to have some power when the grid is down. I plan have enough to charge phones, power radios, power propane appliances and maybe keep me on the internet. I'd like to have enough for a refrigerator, but that starts to get expensive.

I can confirm, I have grid-tied solar panels and have gotten outages in the middle of sunny days which is usually the time my panels are actually supplying surplus power to the grid.

Only way to have juice during power outages in my configuration is to add something like the 7K Tesla battery.

Out of curiosity -- what timeframe are you using to calculate that figure (4-5 cents a kWh)? (I bought a solar system just under a year ago).

Most people would calculate this against the term of the loan. So if you have a 10 year loan vs a 20 year loan, that would change the price. The price is what you actually pay, which in many cases would just be the cost of the loan divided by the amount of energy produced.

Interesting. I would have thought that you do it by (expected) lifetime of the panels. I haven't had my system for a full year yet so I don't have complete numbers, but it seems to be anywhere from .5 (winter) to 1.1 MWh per month.

Assuming I end up generating 8 MWh per year, I'll use that number along with an expected lifetime of, say, 20 years. So: lifetime production of 160 MWh, and my system (after incentives) was something around $18K.

$18,000 / 160,000 kWh = $0.11 / kWh.

If my system lasts 25 years, that goes down to 9 cents per kWh.

(note: I think I have a 10 year loan but will likely pay it off way earlier than that, but it wouldn't factor into this calculation other than interest)

> I would have thought that you do it by (expected) lifetime of the panels.

That would certainly make the most sense to me, and maybe even based on the warranty period, with accounting for replacing failing or underperforming panels after that.

That way, the statement, which sounds too good to be true, "and then the cost drops to zero" doesn't enter into the conversation. Sure, the monthly cashflow cost may be zero for a while, but that might just be accounting trickery.

In some cases it is expected to have less power as time passes by. My manufacturer says to expect a .72% decrease each year, so the total ammount during those 20 years will be smaller.

What I'm wondering is if they're calculating this figure based on peak rated panel power, because if so, that's probably off by a large margin.

Agreed -- that wouldn't make any sense.

With free fuel and falling supply chain costs this is all well and good. Worth noting is that the value of solar electricity goes down fairly sharply as a function of market penetration because of the intermittency. Once you start adding massive chemical electricity storage costs and environmental impacts get more gnarly.

Right. So why add "massive" chemical etc?

People who install solar panels have a choice: ⓐ Restrict their power usage at night, ⓑ install batteries or blah, or ⓒ get power elsewhere. Or a combination.

Many people seem to think that option ⓐ does not exist. Ie. ⓑ+ⓒ has to be massive as you put it, because ⓐ cannot lighten the load.

Ooh, I really like your usage of ⓐ. How do you type those without looking them up on a Unicode chart and copy/pasting?

I use linux/X11, with a compose key and a large repertoire. If I type compose ( 4 ) (that's four keystrokes, in sequence) I get ④, compose ~ e gives me ẽ, which I'm sure some language uses but I've no idea which. http://rant.gulbrandsen.priv.no/xcompose describes my configuration (as it once was).

Again, I'll bring up what nobody seems to mention. With minute-by-minute pricing of electricity, intermittent supply problems can be greatly mitigated without batteries, peaker plants, etc.

On the consumer side, elastic electricity uses can be switched on/off through an app connected to the internet.

Elastic uses are things like charging car batteries, electric water heaters, A/C, refrigerator, etc.

Peaker plants are expensive to run and electricity from them is priced accordingly. If peaker plants are turning on, it's because no cheaper source of electricity was able to meet demand, and demand was inelastic enough to pay the higher prices.

My electric rates are fixed at the same price, 24/7. Trying to maintain this will be very expensive for the grid when faced with fluctuating supply.

Are there any utilities in the US that offer minute-by-minute pricing, or even hour-by-hour? If not, then we cannot conclude that demand is inelastic.

Peaker plants are now more [1] expensive than solar+battery storage. And battery storage that without solar (ie, soaking up night time base-load) are even cheaper.

[1] https://electrek.co/2017/12/13/solar-batteries-to-take-10gw-...

Agreed, I wanted this but the utility can't even do day by day metering.

Reflow batteries have low environmental impact, and are cheaper than lithium ion. The only real downside is that they are about 25% mode bulky for the same energy storage, but for many domestic installations this isn't a big problem.

Many solar plants are now pairing with large scale energy storage facilities too (eg, liquid salt, pump hydro, even some batteries).

I'm curious to know why other storage methods aren't taking off more. I would have assumed that the long-term costs of something like hauling something heavy up a hill (which could easily be a manmade earthworks in flat areas) would be quite a bit lower, on account of the equipment in question having a longer life span than batteries.

The energy density of gravity storage is terrible. But there's still a few plans for it:


> 9,600 tons of rock- and concrete-filled railcars up a 2,000-foot hill.

> $55 million

> The Nevada project has a power capacity of 50 megawatts and can produce 12.5 megawatt-hours of energy.

Using the 4 cents/kWh figure upstream, that's ... $500 of electricity per round trip.

I'm still curious about that project. Energy density is a consideration, but it may not be the only one.

Yeah, as-is, there's no way it can pay for itself. But how much room does this have to grow, and at what cost? If most the cost is down to laying the track and making these custom railroad cars, then maybe there are some economies of scale that could come into play. And then I assume that most of the equipment involved could easily last for decades.

Up-front cost for battery installations may be cheap, but don't lithium ion batteries typically only last a few years? That means you won't be able to amortize the cost of equipment over a very big stretch of time at all.

This makes me wonder, why don't elevators generate electricity on the way down?

Elevators are counterbalanced -- when they are going down, there is a counterweight going up. So there's less energy than you might imagine being generated.

I know, but I can imagine that they would still generate some energy in a busy building. Then again, I don't really know how much energy they consume.

Regenerative braking elevators exist: https://www.asme.org/engineering-topics/articles/elevators/w... has a good summary.

Having lived in a 17-story building with Thyssen-Krupp "green elevators", they're terrible! An elevator would break down and require days of maintenance every couple months, greatly increasing queueing times.

Moving the elevator down would only take potential energy out of the system when the elevator together with its load weighs more than the counterweight.

When it does, you might be able to use it to generate electricity. But I wouldn't be surprised if that isn't done simply because it wouldn't justify the added cost and complexity of building all that extra energy harvesting mechanism into the elevator.

The keyword here is energy density. Batteries are really good when measured in electrical energy stored per volume. Your proposed mechanical potential energy store (lifting mass) cannot compete. The amount of stored energy is aurprisingly low in comparison. I did the math over a decade ago because I had the same idea and the numbers were underwhelming.

Yeah, but energy density isn't really significant. You may get less energy out of a m^3 of water pumped to a reservoir than you get out of a m^3 of battery, but even the smallest reservoir has a usable volume that probably surpasses the volume of all batteries ever built. You can just keep pumping.

No, you can't just keep pumping. To build a reservoir at the scale that you seem to imagine, you depend on locations that provide a suitable topography. Very few such places exist.

A lot of Hydroelectric plants on stable reservoirs can be used as large batteries.

During peak hours they'll increase the flow rate to turbines and during off hours they'll use surplus energy to pump water back into their reservoir.

See https://en.wikipedia.org/wiki/Pumped-storage_hydroelectricit...

Biogenic methane emissions likely make these a non-option at scale.


Or thermal mass storage for AC systems.

What kind of environmental impacts?

Also, what do we do with dead solar panels? Is there a recycling program, or research going into such a thing?

I'm so excited for this news and I feel like these are pretty minor cons though.

Giant batteries aren't great for the environment because they involve heavy mining.

Lithium is harvested from evaporation pools, not mined. If you're referring to cobalt content, there are several researchers working to remove the need for that element from the production process.

And have a one to two decade lifespan, at best. After which you have to dispose of them.

Lithium polymer batteries are highly recyclable, they primarily just have to be unwound, possibly reformed if something structurally changes in the cathode/anode, and the trace elements need to be washed off and processed.

Mea culpa. Thanks for the info!

They are fully recyclable and aren't dangerous to do so.

I'm guessing they meant battery production, which requires the mining of rare-earth elements: https://en.wikipedia.org/wiki/Rare-earth_element#Environment...

Batteries do not use rare-earth elements (as far as I am aware).

You may be thinking of cobalt, which is expensive and mostly mined in the Congo. Reducing the cobalt content of cathodes is a big priority for the industry:


This probably varies by climate. What state are you in? Also, do you get to sell energy back to your power co? And at what rate? Or do you just store it?

The success of solar (and the end of coal) is really a story of the success of natural gas. Gas turbine capacity was so cheap to deploy that most of what's deployed in the past is enough to cover demand, in other words there are very little 'new' gas generators coming online because they're not needed. Solar capacity is much slower to deploy, hence why we will see 'new' capacity added over a long time to come. This is also a good thing, since with gas you still need to provide it was gas and expand the sources of gas and without carbon capture of gas generation it still has co2 emissions. However, gas allowed a massive short term reduction in co2 emissions and displaced coal and will for a long time fill in where solar can't, until we find a way to store energy from coal and that way might actually be synthesised gas.

Gas can also be produced from biological and even synthetic processes. There's been a great deal of research in producing CH4 from CO2 and H2O using surplus electric power from wind, solar, and off-peak nuclear. That means someday all our gas turbines could double as battery discharge machines for a giant chemical battery and the existing gas transport and storage infrastructure could store and transport renewable energy. That's pretty nice from a recovery of sunk cost perspective.

Coal and oil can pretty much only be economically obtained from fossil sources.

While I like the idea of synthetic fuels.

I think combining advanced nuclear power (producing both electricity energy and incredibly good thermal energy) with synthetic fuel (methanol) plants would be far better.

The fact is that the amount of solar and wind you would have to install to cover all intermittent AND large parts of the fuel sector is mostly unrealistic. There is absolutely no scenario the cost of synthetic fuel in large quantities could beat extracted carbon based fuels.

Nuclear is too late. Five years to get permits and five years to build? Perhaps that's optimistic. Hinkley Point C is taking fifteen from announcement to scheduled completion: https://en.wikipedia.org/wiki/Hinkley_Point_C_nuclear_power_... , and that's right next to its two nuclear siblings.

The problem of the current industry are well understood and Hinkley is a perfect example.

However I am talking about something broader then that. Even 5 years ago people would say 'nuclear is to late' solar and wind are gone solve us.

This has been going on since the 1970 when environmentalist turned against nuclear.

Solar and Wind are ALWAYS just around the corner and nuclear is 'not needed' somehow it still tiny.

Nuclear to be effective will need some regulatory and government changes but my real point is that if the whole technical intelligence plus the environmentalist are solar/wind cheerleaders the carbon producers are the one laughing themselves to the bank.

The US and EU have been decommissioning a few older nuclear power plants early because just operating let alone building them is not cost effective.

People talk about base load power like it's a good thing, but it's the opposite demand varies a lot over the day and you want to be able to ramp up and down quickly. You can do this with nuclear, but it pushes costs up even higher.

There is some regional differences and areas with a lot of dispatchable hydro don't care nearly as much. Still nuclear needs to be under 6c/kwh to be viable as base load power or be able to ramp up and down with demand while staying fairly cheap and it can't do either when you look at total lifetime costs.

Most decommissioning in Europe was political.

In the US it it has more to do with how the utilities are regulated. For example, solar/wind get production tax credit when the produce energy that is not needed.

We could go into a long debate about how political management of these market hide many of the costs and so on

> People talk about base load power like it's a good thing, but it's the opposite demand varies a lot over the day and you want to be able to ramp up and down quickly. You can do this with nuclear, but it pushes costs up even higher.

Many modern nuclear plants designs you would build are actually able to load follow. Even in that situation some amount of base-load exists. A nuclear plant actually doesn't cost that much more if you have to run it while demand is low, the fuel cost does not matter in nuclear.

Even worse with solar and wind these things are even worse, you have variability on the demand and the supply side.

> There is some regional differences and areas with a lot of dispatchable hydro don't care nearly as much.

Yeah but that is a fantasy. Hydro is already largely build out all over the world and there is not nearly enough for most dense population centers.

> Still nuclear needs to be under 6c/kwh to be viable as base load power or be able to ramp up and down with demand while staying fairly cheap and it can't do either when you look at total lifetime costs.

6c/kwh is realistic, maybe a little more. But nobody can compete against gas. However against solar/wind and batteries it would easily compete.

Not being perfect does not mean the smart grid/solar/wind/battery approach is cheaper or better in any way. In fact is makes every problem harder, specially in terms of regional differences.

The designing for load following does not directly cost that much more, but it quickly pushes up average kWh costs through the roof while adding a lot of thermal stress.

Nuclear has become a political issue in the US, and these kinds of decisions can be political, not done for economic or safety reasons as one might have hoped.

As an example, the California Public Utilities Commission, decided to shut down Diablo Canyon even though PG&E says in regards to the Diablo Canyon power plant: >...At 2.78 cents per kilowatt-hour, DCPP’s average production costs are lower than all other forms of electricity, but are higher than the national average of 2.19 cents per kilowatt-hour for nuclear power

(Doesn't bother PG&E, the CPUC will let them increase the rates.)


>...Nuclear is too late.

Advocates do a disservice when they ignore the realities of capacity factors. As Bill Gates said in an interview: "…They have this statement that the cost of solar photovoltaic is the same as hydrocarbon’s. And that’s one of those misleadingly meaningless statements. What they mean is that at noon in Arizona, the cost of that kilowatt-hour is the same as a hydrocarbon kilowatt-hour. But it doesn’t come at night, it doesn’t come after the sun hasn’t shone, so the fact that in that one moment you reach parity, so what? The reading public, when they see things like that, they underestimate how hard this thing is. So false solutions like divestment or “Oh, it’s easy to do” hurt our ability to fix the problems. Distinguishing a real solution from a false solution is actually very complicated." https://www.theatlantic.com/magazine/archive/2015/11/we-need...

Gates is investing in 4th gen nuclear and energy storage companies so he is putting his money where his mouth is.

China is investing a lot of money in solar (maybe more than the US), but they don't pretend that will solve all the problems of producing power and we shouldn't either. China plans to have 1400 GW of nuclear power by 2100, so they also realize that decarbonizing energy production will require a multi-faceted approach.


China is currently 2% nuclear - the same ballpark that leads people in this discussion to poo-poo solar as miniscule and irrelevant.

I'd bet real money that the US reaches 6% solar/wind long before China reaches its target of 6% nuclear.

>...China is currently 2% nuclear

Do you have a source for that? According to wikipedia it would look to be at least 50% higher than that:

>...Nuclear power contributed 3% of the total production in 2015, with 170 TWh,[5] and was the fastest-growing electricity source, with 29% growth over 2014.[6] Nuclear generation increased again in 2016 to 213 TWh, a 25% increase,[7] and in 2017 to 246 TWh, a 15% increase.[8] China ranks fourth in the world in total nuclear power capacity installed, and third by nuclear power generated.


>...that leads people in this discussion to poo-poo solar as miniscule and irrelevant.

I haven't seen anyone here saying that solar is "miniscule and irrlevant".

Don't remember... I looked it up the other day as a data point in a discussion of the lack of nuclear power penetration in the market. (I was arguing against the popular idea that the limiting factor on nuclear is the cost of those liberal treehugger regulations, by pointing out that neither Russia nor China have significantly more, despite their disdain for environmentalism.) I was kind of shocked by the numbers, actually. Russia and the US are basically equal in nuclear output percentage-wise, but China lags far behind.

If my numbers were dated even a few years, China is probably doing better, due to current crash efforts to scale up nuclear power. Still, 3% isn't exactly huge market penetration.

As of 2017, wind supplied 6.3% of total US electrical generation and solar supplied 1.3%:


According to an archived version of the page, the shares were 4.7% wind and 0.6% solar in 2015:


Nuclear power can't compete at these prices, even with the new advanced plants.

One day hopefully we'll get fusion, but until then nuclear is dead.

The difference in terms of fusion and fission is not that large. In fact I would argue that any likely fussion plant would have the exact same capx problem that fission now has.

If we can't get fission to work, fusion will not help us.

Compared to carbon based fuel fission is 10^-12 more dense, while fusion is 10^-15. While that is a large change I don't think it matters when competing with wind.

> even with the new advanced plants

I disagree. Once we mass produce nuclear plants the price would be radically reduced, just like in any other mass produced object.

The problem at the moment is nuclear is almost to dense and long running. Its hard getting the scale you need to do that. Specially if every single place you want to put a reactor you get sued 200x by Greenpeace and everybody else.

Compare a nuclear reactor to a rocket engine. SpaceX produces rocket engines for incredibly cheap amounts and he produces 100s a year. A nuclear reactor does not need all that different materials or attention to detail.

The problem with nuclear is the development cost and the regulatory hurdle. If you clear that then you have to find costumers enough to have mass production.

That is why it has not happened and unlikely to on a large scale.

However if nuclear reactors were produced like solar panels and you had global markets, nuclear would trounce solar in terms of price.

I wish i could be a fusion (and new nuclear) believer. But it is just so complicated.

The free market is happy to produce 1M trucks a year. And 10M air conditioning units. And a gazillions flat screen TVs.

How can you possibly compete as the free market works hard to produce just as many wind turbines and solar panels?


The minimum commercially viable size of a power reactor is much larger than the size of a commercially viable wind turbine or PV module. A typical wind turbine today is rated for 2-4 megawatts peak output. A solar module is 0.2-0.4 kilowatts-peak. Commercial light water reactors under construction today are rated for more than 1000 megawatts. Even "small" reactors like NuScale is designing would generate 60 megawatts each.

The large capacity and high capacity factor of nuclear reactors has some significant benefits: once they start operating, they can rapidly displace use of fossil fuels. But it also means that there are many fewer opportunities to iterate on designs. A single large solar project can use a million panels and be constructed over a few years. The millionth panel installed on the site can be significantly improved vs. the first. Most solar manufacturers are tuning module production lines near-continuously and formally introducing new models frequently.

To generate the same amount of power with a nuclear reactor, it would take just a portion of a current Generation III reactor or maybe 2 small modular reactors. To the nearest whole number, there are 0-1 opportunities to upgrade equipment on the fly with a nuclear project vs. the dozens of opportunities with comparable-output utility scale renewable projects.

See https://www.nature.com/articles/d41586-018-05357-w and https://www.economist.com/science-and-technology/2018/06/09/...

"Though that sounds thermodynamically bonkers, such fuel would, from a legal point of view, count as “zero carbon” because making and then using it involves no net release of CO2 into the atmosphere."

It's a way of storing energy, not generating it.

> Coal and oil can pretty much only be economically obtained from fossil sources.

Given that vegetable oil can be used in a Diesel engine after being used for deep fat frying fish and chips, and given we currently over-produce food to the point that obesity is one of our largest health issues, I’m going to have to [citation needed]-you on that.

World yearly vegetable oil production is approximately 200 million metric tons [0], while crude oil is approximately 250 billion metric tons [1][2].

[0] https://www.statista.com/statistics/263978/global-vegetable-...

[1] https://www.statista.com/statistics/265203/global-oil-produc...

[2] http://www.opec.org/library/Annual%20Statistical%20Bulletin/...

[1] says 92.2 million barrels of oil produced per day. [2] says 7.33 barrels / metric ton.

That would put annual crude oil production at roughly

((92.2 * 10 ^ 6) / 7.33) * 365 = 4,591,132,332 (4.59 billion) metric tons.

The implied answer "crude oil production dwarfs vegetable oil production" is still valid even after correcting the crude oil numbers, of course.

Ack, I multiplied instead of divided. Thanks for the catch.

Thanks, that’s a good answer :)


2012, but I don't think the economics have changed significantly. At this point, biodiesel is significantly more expensive than fossil diesel. This would be even more strongly true for synthetic diesel, which is probably economically unviable at any substantial scale.

A gas turbine and a diesel engine are two completely different things. The only thing they share is that they are internal combustion engines.

A gas turbine is basically a jet engine.

I’m not suggesting otherwise. I’m saying that it looks a lot like oil can be produced economically from non-fossil sources.

There's still a lot of excess energy wasted in farming for something like this that you end up with less energy in the oil than was used to make it. See the case of ethanol, which completely failed as a sustainable product and was only propped up by subsidies.

As for capturing CO2, has anyone tried to use it in greenhouses or vertical farms? Some waste heat and abundant CO2 would do wonders for plant growth in colder regions and abundant CO2 would still make plants happy in warmer climates.

Would be complicated with coal (nobody wants radioactive lettuce unless you can enrich it to commercial grade) but I assume gas is much cleaner.

Yes, they do this with greenhouses in the UK. The greenhouse complex has a small gas powerplant, electricity is sold to the grid, waste heat and waste CO2 are used to increase yields.

With the appearance of the "duck curve", utility providers and customers are going have to reckon with the fact that a kWh of solar electricity at the noon hour is not worth the same as a kWh of solar electricity at 5pm.

So long, dinosaur coal plants. My lungs are looking forward to solar eating the rest of your lunch.

There is a coal powered plant in Eastern Montana that is having troubles.

"Data processing and blockchain interests are the latest entities to court the ailing Hardin Generating Station, a 109-megawatt coal-fired power plant that is seeking a buyer to avoid closure." - https://www.apnews.com/6a577c7d0dc1488b9481d03685950e96

Great, now we can watch train loads of coal go into a building and bitcoins come out! Yeah, that is the smell of progress.

Here's pie-in-the-sky idea, what about building high-voltage transmission lines across the continent?

In the case of US, New York at 8PM would get electricity from California at 4PM. There would be some losses from transmitting power for such large distance, but shouldn't it be better than just wasting energy or storing energy even less efficiently?

Its not great. Even before solar the advantage of a cross continental TL were there (namely ironing out the demand curve, hydro storage, ect).

To limit resistive losses you have to go to really high V -> massive pylons (there were some built behind the iron curtain).

But you still have capacitive coupling (and reflections after the TL is larger than $\lambda$/10 -> You need to go to very low f or even DC

But high V. DC is hard! and low f AC is massive.

Ultimately it hasn't been done because it hasn't been worthwhile.

HVDC is harder than AC transmission, but isn't it a solved problem?

"In July 2016, ABB Group received a contract in China to build an ultrahigh-voltage direct-current (UHVDC) land link with a 1100 kV voltage, a 3,000 km (1,900 mi) length and 12 GW of power, setting world records for highest voltage, longest distance, and largest transmission capacity."

(From https://en.wikipedia.org/wiki/High-voltage_direct_current)

Can you expend on some of these, or give a reference? Specifically, why does going to high voltage limit resistive losses? Why does really high V mean massive pylons? Why is low frequency AC massive?

P = V * I [1], and V = I * R [2], so P = I^2 * R.

So to minimize power loss in the cables, you want to minimize the current. In order to transfer the same amount of power, you need to increase the voltage.

Not sure about the pylon sizes.

[1]: https://en.wikipedia.org/wiki/Electric_power

[2]: https://en.wikipedia.org/wiki/Ohm%27s_law

NZ has been running an 600 km HVDC line for decades for exactly the same reason - lots of hydropower in the south, lots of demand in the north. difficult? certainly. but it's definitely a solved problem.


Seems like it's not such a crazy (pie in sky) idea after all: https://www.technologyreview.com/s/609766/how-to-get-wyoming...

Isn’t this a 75% solved problem, because power generation and use were already out of sync due to people not working at night and a significant fraction of power generation being fixed-output such as nuclear?

No, power generation and use are definitely not out of sync today. Generation is very very tightly synchronized to usage and has to be for the grid to work at all without storage. Quite a bit of effort goes into managing generation output in real time to maintain proper voltage and frequency.

The solution is storage, but at current pricing the amount required for a full renewable grid would be obscenely expensive. Luckily battery prices have been dropping quickly for years and continue to do so.

Grid battery storage is a pretty mediocre solution to variability in supply from renewables except in a few cases (e.g. where pumped water storage is readily available). It's expensive and inefficient.

Market driven demand side flexibility is not very well tapped, doesn't require any technological leaps and there's a lot of room for improvement there. For example:

* Aluminum smelting plants that vary their usage depending on spot cost (actually already happening in Germany).

* Smart electric storage heaters that turn on when electricity is cheap.

* Smart electric vehicle chargers that take electricity spot price and current charge levels in to account when deciding when and how much to charge.

* Hell, even bitcoin miners.

When there are millions of EVs on the road charging by day at workplaces and discharging at night at consumer homes, I don’t think they’ll be observed as mediocre solutions.

And killing the car's battery with extra cycles and leaving the commuter with an uncharged battery in the morning.

Two things:

1. Cars have very very large batteries. You'd be hard-pressed to use more than a third of the capacity of a high-end Tesla Model S over night (100 KWh!). Sure, maybe you're driving a Nissan Leaf and its battery is only a third of that capacity, but in a few years 100 KWh might actually be common. What if you want your car battery to be at 100% in the morning so you can go on a roadtrip? Just make sure your car is aware of it.

2. There's going to be a marked for used batteries from old, retired electric cars. You might only need 20-40 KWh of remaining capacity to live comfortably in a fully off-the-grid solar setup.

Tesla battery longevity data shows this isn’t a concern. You’re not going to deep cycle the battery every night, and while I can’t speak to other auto manufactures, Tesla will be increasing their pack sizes, allowing for more “spare capacity “.

Or we’re going to subsidize residential storage (such as the 50k home project in Australia).

AFAIK it depends on the location.

For example, where it's hot, people are running aircons during the day. Where the weather is cold, you need to run heating, but (1) often that's not electric, and (2) heat stores relatively well (in e.g. hot water tanks). In temperate climate, the uncomfortable power surge happens at ~6pm, when people get home from work and start turning on the light and running home appliances (stove to cook dinner, opening the fridge which must cool, etc).

The main power draw during the day are in the morning and at the end of the day, not in the middle of the day when the solar is most efficient. That's why it's called a duck curve.

Exactly - even in summer, I recall reading that utilities have been begging for power consumption.

With Power pack like products I also wonder if there's an opportunity to 'level wear' the energy demand programmatically.

Some areas in Europe do offer spot pricing (~hourly interval IIRC) for electricity, even for consumers. I don't have such a contract, so I'm not sure how good the pricing APIs are for autoscaling your consumption.

Several utilities in the states offer hourly rates. Most use human behavior inputs (sms messages) to drive the changes but others actually deliver the spot prices via smart meters.

A less smart version already exists. In the UK, some domestic energy is supplied at different costs at different times of day. This is optional and typically used for heating, but means that night time energy is cheap compared to daytime energy.

of course, I got caught in my technological views, but the price ranges are made for this entirely

A rather biased source for sure but last years first solar analyst day presentation had an interesting remark regarding the duck curve. See page 3 of http://investor.firstsolar.com/static-files/2c0866ad-c821-49...

But yes it is going to be harder and harder to compete with solar in a supply priced market.

Is that file available anywhere else? I get "Access Denied".

It is the technology update from last (2017) first-solar analyst day. Page 4. Key part for the duck curve discussion is that there is a big duck curve on a low load day in May but no such duck curve on a high load day in August.

Then they talk about how their solar is starting to be dispatchable (market wise) by curtailing production. I.e. if they commit to sell 80% of their production then 20% of it becomes available as dispatch able power. That only makes sense if the solar power is cheap enough. Which it is does in some situations in some markets.

There's no content on page 3...

Sorry page 4.

> So long, dinosaur coal plants. My lungs are looking forward to solar eating the rest of your lunch.

Absolutely. The coal industry, like any industry, will obviously attempt to hang on as long as it can. However the US is down to a point where there are only ~50,000 coal mining jobs left (vs 175,000 in 1985). Solar, wind and natural gas can easily replace those.

There's no way coal pulls out of this decline (EIA first quarter updated figures):


It'll take a little longer than you think.

Further, you're referencing a coal job figure, which is meaningless in this context. Coal production in the US over the past decade was higher than ever, by quite a large margin too, it's just that humans are no longer necessary for much of it.

I'm not a fan but these are the facts. Fortunately, production is now dropping worldwide, but it really is early days, it's still way up compared to 10, 20 or 30 years ago, and it certainly isn't because of renewables. It's a gas-driven change.

> Coal production in the US over the past decade was higher than ever, by quite a large margin too

No it wasn't. It peaked in 2008 and is down quite a bit since then. 2016 production dropped 18% over 2015 and was the lowest since 1979. And despite the current administrations best efforts, 2017 production increased ~6% over 2016.

I see two to three coal trains a day come through my town, with at least 40 hopper cars of coal.

The coal is still going somewhere. I assume overseas, but unfortunately (in this instance), as people of earth, we all share the same atmosphere.

Assuming you live in the US, the US doesn't export much coal. It's going to remaining coal plants, probably. But these will eventually be decommissioned, and it's highly unlikely that new ones will be built.

US producers are trying to partially compensate for declining domestic demand with exports through new export facilities in California and Washington. Unfortunately, exporters may be able to force facility approvals through against local and state wishes:



Coal still makes up about a third of US electrical generation. It’s declining, but it will take a long time to disappear.

It's not really about coal at this point the war is between solar/wind and natural gas.

While I agree with that sentiment, when over a quarter of US electricity generation is derived from coal, it's still clearly about competing with coal to drive it into the ground. Between the US, Europe, Japan, China and India, there's about $23 trillion in economic activity that depends on coal. Certainly all the generation growth and replacement going forward is contained in that war between renewables and NG.

> competing with coal to drive it into the ground.

Good choice of words, in that what we need to do with both coal and oil is leave it in the ground; which means that the people who've invested in its extraction need to be overcome somehow. Whether that's buying them out, making it uneconomic, or something more coercive.

I think my point is the economics and uncertainty for coal is so bad now it'll disappear even with idiots like the Trump administration trying to subsidize it. Coal usage in the US is down 1/3 in ten years and that's going to accelerate.

A dream would be to have onshore and offshore windmills where it makes sense, solar panels on most residential and commercial roofs, panels serving as roofs for car parking lots, hydro power generators where water flow needs to be regulated, and to supported such a renewable power generation - connected micro grids with local battery warehouses to serve as a power buffer.

Some centralised govt provided services could be run on "excess power" such as pumping water to a overhead storage for a local community, or powering a water purification or sewage cleaning plant to run during periods of excess power.

Now, I need to wake up and go get a coffee :-)

> solar panels on most residential and commercial roofs

If one's aim is progressing along the Kardashev scale [1], then the single metric to optimize is the cost, including of externalities, of energy. Rooftop solar seems doesn't seem to scale as efficiently, both in measured and externalized environmental costs (e.g. installer safety and carbon consumption), as utility solar.

[1] https://en.wikipedia.org/wiki/Kardashev_scale

>If one's aim is progressing along the Kardashev scale

I doubt most folks who put solar panels on their roof are doing it to advance a hundred thousand year grand strategy for galactic scale energy consumption.

I think it's mostly stuff like minimizing carbon footprint and saving money over the 20 year lifetime of a project.

> saving money over the 20 year lifetime of a project

Practical point being, a future where everyone has rooftop solar is a future where everyone is spending more on energy than the alternate future of utility-scale solar. If we want to save money, the cheapest solution, i.e. utility-scale solar, is the way to go.

While I understand what you're getting at here, I see little evidence to suggest utility controlled generation will ever be cheaper.

People are installing roof-top solar now because they're able to seeing savings in the form of reduced electricity bills now.

Practical point being, if you tell people: "don't install roof-top solar now, keep paying outrageous electricity prices (I'm in Australia, see) and hope that one day the electricity retailers pass on the savings when they can buy wholesale solar / wind for next to nothing" you can expect to be laughed at.

Can't speak for Australia, but at least in the US, a big part of the reason rooftop solar ends up looking like such a savings is net electric metering, where power companies buy surplus power from solar panel owners at the same rates that they sell those users power at night. The tricky thing is that the price users pay for power doesn't just include generation costs, but also transit, grid maintenance, etc., which rooftop solar owners benefit from even though they no longer pay for it -- these costs end up being borne by other rate-payers, who effectively cross-subsidize rooftop owners' freeloading. In the early days of solar, there was a good public policy argument to be made here: solar power is a net societal good (air quality, carbon footprint, creation of solar economies of scale that drive prices down for everyone, etc.), and power companies weren't going to do solar on their own.

Now, though (again, at least in the US), utilities are making massive investments in grid solar, and it's dramatically cheaper than rooftop solar. Like: a quarter the cost per kilowatt hour of generation capacity. So now I think it's sensible to ask: if we're going to ask ratepayers to subsidize solar generation, does it really make sense for us to spend their subsidy on a way of doing solar that's objectively much more expensive just because it saves a few specific users money? They could be getting 4x the bang for their buck if we spent that subsidy on grid-scale instead, and then the economics of rooftop solar mostly go to shit (as well they probably should -- it's needlessly expensive).

In Australia solar is cheaper than the grid even taking account of battery storage.

I think it certainly could be cheaper over the time scales relevant for Kardashev scale. But we have plenty of time to worry about that later.

But we need to figure out how to survive the next couple of centuries before we start worrying about efficiencies that will matter in tens of thousands of years, and rooftop solar certainly helps in the immediate term.

Utility scale solar is definitely cheaper. NREL puts residential at $2.80/watt vs $1.03 for utility scale[0]. Its only cheaper for consumers because of subsidies and net metering.

[0] https://www.nrel.gov/news/press/2017/nrel-report-utility-sca...

edit: Missed the part where you are in Australia, I was giving numbers for the US. I believe its true the residential numbers are cheaper in Australia due to lower soft costs, its still true that utility scale has a cost advantage because of labor cost savings.

Different entities have different costs.

Depending on the installation, maintenance, and component costs of different parts of a system, it may end up making a lot of sense to massively downsize the grid and make most single family homes into islands.

Electrical grids are hugely expensive, and even if the raw cost of utility scale solar installs is lower than residential, the delivery is killer.

You’re optimizing for the wrong outcome. Cost is only one factor. There’s also being immune from utility rate increases, on-site perpetual backup power generation, and reduction of transmission and distribution costs.

Disclaimer: I have rooftop solar.

Is it really cheaper? There are some economies of scale, but how much? Rooftop solar has zero land cost and low transmission costs.

Yes, over twice as cheap. According to NREL, residential is $2.80/watt vs $1.03/watt for utility scale.


That doesn’t seem to account for the two factors I mentioned, although I certainly wouldn’t expect them to make up that difference.

I haven't read the NREL report, but my father was involved with them for 10 years as an investor. For utility scale, they can use thin-film panels which are more efficient, but more costly upfront (FirstSolar is the big player).

I think humans can progress on that scale, but I also think we need to do it the human way. That is, the scenic route if you will, the longer way. If we can just nibble our way to full coverage (of roof tops in this case) faster than we put up new roofs, I think we'll be fine. Utility solar will follow eventually.

Also, and this also ties in to mankinds irrationality - say there is some kind disaster or bombing campaign. I fully believe roof top, i.e. decentralized power generation must be more robust than utility solar.

Admittedly naive ask. I hear this argument that residential solar panels are not very cost-effective. A former colleague once told me that after Government subsidies, his break-even point for solar investment was 12 years. I didnt ask him at that time. I'm assuming he was referring to energy buyback from his utility provider for his math. Isn't that against their business incentive to support residential solar?

If one has a house, and covers the roof with self-purchased solar panels, and hooks them up to a large enough battery-box, wouldn't it just be "I saved the cost I would have paid as my utility provider's bill"? That way the return on investment is more manageable and is without being held at the mercy of the power companies?

>Isn't that against their business incentive to support residential solar?

Yes, but that may not matter.

I have a set of solar panels. They cost about £5k. They produce up to 25kWh/day: https://flatline.org.uk/daystats.html

I'm paid £0.138 for every kWh that is generated, regardless of whether I use it or not, and I don't have net metering. This results in a cheque every few months for about £500 a year, plus the saving of a difficult-to-meter amount of electricity used locally during the day. This gives me a ten year payback (or sooner).

Entertainingly because of the way feed-in tariffs work in the UK, they could be paid by a totally different provider to the one billing me for grid electricity.

Also, these calculations seem to assume electricity costs will remain constant, and not increase.

Also assumes the cost won't go down. I fear utilities will start jacking up monthly customer charges and keep rates low to avoid paying solar customers for generation.

> one's aim

Who's the "one" here? I mean, that's simply not how our current civilisation works in terms of "aims" in any collective sense. We're a long way from the central planning comittee erecting a Dyson sphere.

Realistically, the objective of various political bodies is to avoid global warming catastrophe in the medium term, while taking into account the irreducible political constraints and limits to planning. Rooftop solar through subsidies is easier, because suddenly you've recruited an army of people to handle the permitting and the process is simpler. Whereas to build utility-scale solar you need to spend a year or more arguing over it before breaking ground.

Rooftop solar cannot work in dense multistorey apartments. Its at best a rural suburban solution

Dense multi-story apartments already are far more efficient than the same density spread over a suburban sprawl.

Which is why diversity in the energy portfolio is a good idea.

That is more of a idealistic fantasy world then a reality.

Roof solar and smart grid are incredibly expensive.

Why not just look at countries who have already solved the problem. France, Switzerland and Sweden all have essentially zero carbon in energy production. They all use water and nuclear energy.

For the US, you can just start replacing goal with nuclear, without the massive cost of the fundamental transformation to the whole energy system and grid.

Is the goal to create an idealistic distributed energy system or is the goal to effectively reduce carbon?

Every time one of these posts go up, I hear the same arguments about nuclear power. The problem is that the utility companies in the US who make the decision about what kind of power plant to build are all backing away from nuclear. And they know more about this than you or I as it is their business to know.

These companies have alot of good reasons for doing so, but they all boil down to the overall cost of nuclear being more expensive than any other alternative. Some of this is regulatory, and you can possibly loosen those, but alot of it just has to do with a lack of suitable sites and current plant designs which are based around the idea of providing huge amounts of capacity to large population centers. There are only a few types of utility companies that fit the bill in the US, and they already got one in most cases.

Theoretically, you could have designs that provide smaller amounts of power distributed over a wider area, but then you have security concerns, and no matter what design you use, they all require access to a large body of water for cooling purposes, which means you can only make them so small in terms of area.

While solar is not able to provide as much energy per plant, it can be scaled to fit the needs of the utility company and doesn't require as much in the way of extra infrastructure and has fewer regulations. So it will always be used by a broader range of electricity providers even if we were to work to replace some coal with equivalent nuclear capacity.

> just start replacing goal with nuclear

That will take a decade or more before any energy is produced, even assuming a crash programme to build them starting now. Reducing carbon needs to be done in this decade for the Paris agreement targets to be met.

Great. And solar/wind have absolutely no change of doing so. Germany has spent years and years with tons of money and their progress is very slow.

Had they done nuclear they would be about done by now.

Do you have other energy sources that I am not aware of?

Is safe storage of the radioactive waste from nuclear plants a solved problem?

Even if the current fears about nuclear are overblown (and I'm not convinced that they are) what do you do about waste that remains dangerous for many human lifetimes?

> Is safe storage of the radioactive waste from nuclear plants a solved problem?

Depends on what you mean by 'solved'.

Its not solved in the sense that there is a political solution and agreement.

It is solved however in so far as that all scientists and technician agree that there is a simply solution and that the money for that has already been accumulated by fees payed on nuclear energy.

We are just now again seeing some progress on this front but sadly this topic has been covered in mountains of shit.

> Even if the current fears about nuclear are overblown (and I'm not convinced that they are) what do you do about waste that remains dangerous for many human lifetimes?

Actually the whole way we talk about this is wrong.

Nuclear 'waste' is not waste in any meaningful sense. It actually contains highly valuable materials. The waste majority of the 'waste' we have now is fuel that can be used in reactors (if we ever manage to build new reactors). Other parts of it can be used for nuclear batteries that we use to explore space, other parts can be used for medical isotopes and the list goes on.

There are geological places that have no moved for millions of years and will not for millions of years as far as we can tell.

In order for it to endanger lives you would have to come up with incredibly complex story. Nobody has ever died from nuclear waste and its very unlikely that anybody ever will.

Compared to solar or coal the waste is far smaller. The output of coal plants is ironically more radioactive then a nuclear plant and ash pile is worse then all nuclear waste combined.

Solar leaves behind mountains of electronic waste the then go to childeren in Africa that take them apart and get poisoned.

Nuclear has the smallest amount of population or waste compared to any all other technologies. It is by far the safest for human in both cases.

It is also the saves form of energy production (including all nuclear disasters) in terms of human lives.

> Nobody has ever died from nuclear waste and its very unlikely that anybody ever will.

This is a very dubious claim given the former Soviet attitude to nuclear leaks.

The 'unlikely anyone will' is quite likely, though, given how much more attention is given to nuclear waste compared to other highly dangerous cemicals.

Nuclear scare is the case of zero-risk bias, as by discarding nuclear people are forced to use co2-emitting methids for base load, making global warming and pollution casualties far greater then nuclear ever could. [1]

[1] https://www.statista.com/chart/13994/global-deaths-per-energ...

That could be true but even then it would likely be because of long term higher statistical risk and a lack of medical care.

Several points here:

1. Most waste from nuclear plants is contained, unlike coal/gas. 2. The amount of the waste itself is miniscule, the amount of uranium needed to produce all the energy you will ever use throught your time on this planet will fit in a can of coke [1] - for comparison - as shown for example in [2], it would take 20 tonnes of coal pulverized into atmosphere to get the same amount of energy. 3. Being highly radioactive means decaying fastest, so most dangerous are also the short-lived. 4. Unlike coal/gas, there are many projects to use the waste from nuclear plants to use spent fuel from one kind of plant into another. [3]

[1] https://bravenewclimate.com/2010/04/22/ifr-fad-4/ [2] https://signatureelectric.ca/blog/how-much-energy-does-the-a... [3] https://www.fastcompany.com/3043099/this-nuclear-reactor-eat...

It's a beautiful dream and I hope it happens

I'm curious if anyone has gone the Tesla Solar Roof route?


Here's what I'm looking forward to: a world where energy<>pollution.

This whole period where energy production as our greatest environmental problem has promoted the wrong type of thinking. It's Energy austerity, the need to take steps back, slow or reverse progress.

Since technology, industry and energy are so closely coupled, this has almost made futuristic thinking politically incorrect, or at least unpalatable. I want to get back to a future that's futuristic, where we can build 100X bigger, desalinate sweetwater rivers, have floating stuff all around us....

Basically, I want to get back to the old curve, where we make more energy every year and that's a good thing.

Non of this thinking is even futuristic. It was futuristic in 50s.

Look at France. They solved this problems in the 70/80 with technology from the 50/60.

The problem is utter stupidity of both the population and politics.

There is absolutely no problem. Nuclear energy is utterly inexhaustible for the next 1000 years, it uses almost no resources (in terms of land, steel, water) it causes absolutely no pollution at all.


Why not copy what clearly works? Why does everybody invent something new when we invented something 70 years ago.

We would finally have the energy to do things like desalinate sweetwater and synthetic fuels.

Nuclear is too expensive, the French option is massive centralised planning for scale, combined with subsidy to encourage an export market, neither of which are palatable in much of the developed world. There’s also the proliferation risk, in combination it’s just not going to do what we need it to do.

France did it with 50/60 technology in 70. France did it fairly centralized but there are other options to get to a similar place.

Today its far easier to mass produce and scale. Modern reactors are better design for this.

Nuclear growth in the developed world would actually be possible. That is actually what China is planning to do. They want to sell reactors and get fuel contracts to control people infrastructure.

Most experts agree that civil nuclear power actually helps proliferation.

My problem is this. You read news article after article about how horrible and disastrous climate change already is and will be even worse very soon. WE NEED TO ACT NOW.

When you then go to the same newspapers and people, they are talking about how solar is coming along nicely and with government support in 20 years it be relevant.

Nuclear could have been adopted (or continued) 30 years ago. Carbon would never even have been a large problem. Only having to change transportation would have been far easier once you have non-carbon energy.

Nuclear construction is great for energy supply throughput, terrible for latency.

Today its far easier to mass produce and scale. Modern reactors are better design for this.

This is not supported empirically. France's Generation III reactor design, the EPR, is proving slower to build than Generation II reactors in the 1970s. The first French EPR, Flamanville 3, is currently on month 102 of construction, when it was originally supposed to be complete after 54 months.

The two Generation III AP1000 reactors under construction in the USA are now expected to be completed in 2021 and 2022 after 9 years of construction. They were originally supposed to be completed in 2016 and 2017. Two others were cancelled after years and billions of dollars were spent because they could not meet schedules.

Note that EPR and AP1000 reactors are also under construction in China and they are years late there too. These dreadful mis-estimates can't be laid solely on Western laws or environmental activists.

Before construction actually started on EPR and AP1000 reactors in the West, project boosters were claiming that Generation III reactors would be safer, have predictable construction schedules, and be affordable. None of those boasts have been affirmed yet. Nuclear boosters sometimes excuse this with "the first of a kind is always slow." But then talk of "fast" nuclear construction is even less realistic. It takes a decade to build the first of a kind. Then if the debugged construction process gets down to a brisk 4.5 years per reactor, like China has managed, it's about 15 years before you see any kind of fast decarbonization progress from a commitment to build a new generation of nuclear reactors.

I think that the US should keep reactor construction alive so that the know-how isn't entirely lost again, and certainly should not shut down already-built reactors that are still in good working order. But that still wouldn't be a fast fix for cutting emissions.

Nuclear gets cheaper if you actually have production of many units. This is pretty clear in the literature.

If the US planned to replace all coal plants with AP1000 in the next 15 years it would be a totally different process.

Look at the prices when France or the US were building many reactors.

Also know-days you can already order thinks like Pebble Bed reactors from China or Molten Salt Rector from Canada.

> it's about 15 years before you see any kind of fast decarbonization progress from a commitment to build a new generation of nuclear reactors.

And if you look at all the different plans this is still about the best.

The 2050 all renewable plans are FAR more unrealistic then any comparable plan with nuclear.

Once you get the build time to 5 years its becomes economical and that is very realistic once you plan for more then a few identical reactors.

> If the US planned to replace all coal plants with AP1000 in the next 15 years it would be a totally different process.

Which goes back to the need for centralized planning, which is unpalatable in most Western countries.

> Also know-days you can already order thinks like Pebble Bed reactors from China or Molten Salt Rector from Canada.

You make it sound like you're ordering them from a catalogue. Those are research projects, hoping to eventually produce a commercial project.

France is also smaller than Texas and has a much larger economy with twice as many people. And France was rebuilding alot of its infrastructure from WWII which gave it an advantage in terms of cost compared to the situation the U.S. was in at that time. The U.S. is much more spread out and geographically diverse than any of the examples you've given and there is no "one size fits all" solution for that reason.

Nuclear power is good for dense populations, but it is much less cost effective for low density places which includes half or more of the US geographic area. Which means you're left with subsidies specifically for nuclear plants in those areas, or something like a carbon tax to make them competitive, which would also benefit solar and other alternatives that are a better fit to begin with.

I’m very very pro nuclear, but you’re sort of bullshitting here. No pollution? Most people would consider nuclear waste to be polluting, and nothing is perfect, sometimes you get a Fukushima. It also pollutes down the line when you’remining Uranium and processing fuel, and for now even with perfect safety that’s unavoidable.

It’s still a much better choice than anything else, and it pollutes less, and modern designs can be very safe. We’d actually be safer, since we could replace older, unsafe designs with modern ones. We still need a political solution to the waste problem though, and I have no idea how that can be done.

> Most people would consider nuclear waste to be polluting

Then 'most people' don't know what pollution means. Nuclear 'waste' is a controlled output of a process that can and is controlled to not effect the environment in any way.

In terms of nuclear arguing what 'most people' believe is utterly pointless because 'most people' don't understand even the basics of nuclear physics.

> and nothing is perfect, sometimes you get a Fukushima

Nothing is perfect. But somehow people don't point to all the deaths that happen on solar installations and so on.

The fact is that nuclear is the safest form of energy.

People are somewhat irrational in nature and fear rare, but spectacular events more than the mundane. This is the number one problem with nuclear that I see it: no other power source has had anything on the scale of Chernobyl. The only thing that I think offhand could come close as far as potential deaths is a hydro dam collapse.

The reality is, we are all human, the world loves to throw unexpected surprises at us (such as the tsunami that helped cause Fukushima). So a Level 7 event is always a small probability. Even if it is much less of a probability than fossil electricity accidents and the issues from pollution overall are worse for fossil fuel, it's a hard sell because of the ways humans are wired.

It also doesn't help that the technology is mysterious to a lot of people as you mention. "Flammable material burns, makes energy" is somewhat intuitive to most people. Nuclear is not intuitive at all. Just think at how Hollywood tends to depict nuclear in the popular lore. :)

At any rate, my personal reason I think nuclear will never gain much traction in this modern world is that one of those "potential surprises" is terrorism. Nuclear plants are an acknowledged target. So at the very least, you have to add costs to safeguard the structure and really beef up the security around the plant these days. IMHO this would make any big central nuclear power solution quite a bit much less attractive.

> People are somewhat irrational in nature and fear rare, but spectacular events more than the mundane. This is the number one problem with nuclear that I see it: no other power source has had anything on the scale of Chernobyl. The only thing that I think offhand could come close as far as potential deaths is a hydro dam collapse.

You don't need to think of the potential. We have already had power generation accidents far worse then Chernobyl. The Banqiao and Shimantan Dam collapse in China killed 170,000 people, and destroyed the homes of 11 million more.

And if you want to rank energy safety by KWH generated, you can look at:


We've also had a pipeline explosion that killed a thousand people, we have five Chernobyls every year thanks to the poor safety record of coal mining... The list goes on.

That’s a very selective response, but ok, I can do that too.

In terms of nuclear arguing what 'most people' believe is utterly pointless because 'most people' don't understand even the basics of nuclear physics.

It’s The point, because it’s their representatives that have stopped all progress in the space for decades. We’re not getting our nuclear power unless they change their minds en masse, period.

I agree with that.

But that's the original point why I wrote my main comment in this thread. The solution is there but the media and the liberal elites rather spend their time promoting solar and batteries.

Can be safe and will be safe, if the world at large commits to massive expansion of nuclear fuel production are two different things. Theory and practice. Modern IT systems also guarantee full recovery from any data loss. That doesn't mean data loss never happens.

In reality, we will have fuckups. More worryingly, nuclear as the fuel of the future has security implications.

The fact that the problem is theoretically solvable is important.

I'm with you, the answer depends on the specifics.

While 100% can never happen nuclear is a well know technology with a record of safety better then anything else.

In any category you measure environment or safety.

Modern reactors that we would build in the future are a fair bit more secure.

> More worryingly, nuclear as the fuel of the future has security implications.

While true in a global context most countries that use massive amounts of carbon already have some sort of nuclear industry anyway.

1. If only energy were the only problem. Extinctions, resource depletion, overcrowding, etc. I like futuristic technologies. I also like kids playing on grass, not astroturf, however space-age.

2. Consuming less doesn't require austere or lower standard of living. Most Americans can reduce their energy and other consumption probably by 75% and improve their standards of living.

My standard of living has improved since acting to consume less and I'm not special.

can you write a guide on how to improve your standard of living while reducing consumption?

I would think that reducing consumption while maintaining standard of living would have high initial capital costs..

I would have used quality of life, not standard of living.

There are material solutions that have high initial capital costs, but there's also a ton of behavioral or lifestyle changes you can make.

One of the biggest is meat consumption. A calorie of meat has something like 10x the energy embedded it in as a calorie of vegetables. I'm not a vegetarian by any stretch of the imagination -- I just love ribs too much -- but I have significantly cut the amount of meat I consume to a few times a week. Helps you get more creative with cooking, opens up new tastes, and frankly, costs less.

Transportation is the other big one. I've started biking for my commute and it makes me feel great... I've lost weight and my commute just feels better than being stuck in a car or bus every day. That one could arguably be a lower standard of living (if you measure standard of living by the comfiness of your car) but in my opinion, definitely a higher quality of life (I'd rather be moving than stuck in traffic or crammed into a standingroom-only bus).

Thanks. Those are both good examples. And I agree quality of life is better metric than standard of living.

The amount of energy sent our way from the sun every day is more than enough (I half recall a fact that one day of sunshine has the same energy as all the fossil fuels left on earth, someone correct me).

I think with solar getting cheaper all the time the challenge is going to become storing energy cleanly and distributing it efficiently.

Okay, in December 2017, the US produced 345,939 GWH of electricity, of which 1,476 was solar.


If you look at the best spots in the US in December, they seem to get about 4,500 watt hours per square meter per day.


So that means we'd need 76,875 square kilometers of solar to match the current generation of the US, not counting losses from storage.

That's between the size of South Carolina, and West Virginia, just coated with solar panels.

Yes, that's huge. The big numbers can look overwhelming. Huge land-use changes that we've already made would look overwhelming if considered as a single monolithic project. Solar facility construction isn't going to be a single monolithic project any more than the existing American biofuel or fossil fuel systems were.


2016 numbers: about 93.6 million acres planted for corn. About 5.2 billion bushels out of 14.4 billion bushels of corn yield went to corn ethanol. If harvest and land area are proportionate, that's about 33.7 million acres dedicated to corn ethanol [1].

33.7 million acres is about 136,000 km^2. The areal energy yield from biofuels is paltry compared to what you can get from solar PV. Manufacturing/installing solar PV is more resource-intense than planting corn. But it's low-maintenance once installed. It doesn't require arable land, weeding, irrigation, fertilizer, or harvesting.

[1] Yes, the residual material left after fermentation is also used as animal feed. But the conversion efficiency is low; you're not adding many calories per acre to the human food supply when corn is fermented and leftovers go to cattle.

The good thing about solar is it need not occupy new land. It can be installed on land already "occupied" by buildings.

According to NREL, there's enough existing roof space for 1,432 terawatt-hours of annual energy generation from solar, which would be 38.6% of our annual generation.


Or 70% of Nevada.

I for one welcome our paving-the-deserts-with-solar-panels robot overlords.

It doesn’t help to capture all of that energy if we can’t store and transmit it efficiently.

For sure!

And I think that a massive increase of storage will make transmission much easier, by allowing us to time-shift transmission—at the expense of efficiency.

1. generate solar power during the day in Nevada and store it nearby

2. trickle it cross-country at night—when the grid is under-utilized—to distributed storage facilities, where demand will be higher the next day

3. use it the next day, or store and forward again

That way, long-distance transmission won't need the same massive upgrades—high-temp superconductors! megatons of copper! 7-figure transformers!—as if we had to use the solar power right away.

And it's going to get even bigger.

Just yesterday they announced a 2.49kW/h deal for solar in Arizona[1]. That's cheaper than the "low" price of wind power (which is the cheapest source).

[1] https://www.pv-magazine.com/2018/06/11/lowest-approved-solar...

[2] https://en.wikipedia.org/wiki/Cost_of_electricity_by_source#...

Wow, that's cheap.

To give some perspective, an average cyclist can sustain about 100 watts, so generating a kilowatt hour takes ten hours. You'd then need to eat about $5 worth of food to compensate for the exercise.

So buying this electric is 200x cheaper than generating it yourself with a bicycle!

At this point I'd like to start a flame war about the CO2 impact of cycling ;-)

To clarify, that should be 2.49 cents (USD cents) per kilowatt-hour.

And today, there's a new low in Nevada: $0.002375/kWh on a 25 year contract.


To put things in perspective. The amount of solar power produced in the US is roughly the amount estimated to be used in indoor cannabis growing. https://motherboard.vice.com/en_us/article/9a3bd8/growing-ma...

Is there a DIY on powering a whole house using solar panels that are independent of any company?

Sure there is, scale it yourself.

I like Misouri Wind and Solar (https://mwands.com). I've also looked into Wholesale Solar (https://www.wholesalesolar.com/) which focuses on DIY installation and design.

I wonder how will these new green energy sources affect the petrodollar once electric means of transportation start becoming commonplace. Will it have a big impact?

> the petrodollar

Petrodollar hypotheses get the dollar's hegemony in international trade backwards. Petrodollars (i.e. oil producers' surplus revenues invested in U.S. dollar denominated instruments [1]) don't cause the hegemony. The United States being (a) the only advanced economy not bombed to the 19th century post-WWII and (b) the world's largest single consumer market [2] created and create international holders of U.S. dollars. (In the initial case, thanks to Bretton Woods. In the recurring case, since Americans pay with U.S. dollars.) Those holders, in turn, drive demand for dollar-denominated investments.

The interaction between these processes, international consumption by Americans and the reinvestment of international dollars, drives economies of scale in the American financial system. (It is why, until very recently, it was cheaper and faster to send U.S. dollars between two European capitals than initiate a Euro-denominated transfer.)

TL; DR Every watt the U.S. generates from solar is a watt of (a) cheap, domestic energy and (b) petroleum the U.S. has the choice to export. Cheap and plentiful energy drives economies; economies, ultimately, drive demand for their currencies.

[1] https://en.wikipedia.org/wiki/Petrocurrency

[2] https://en.wikipedia.org/wiki/List_of_largest_consumer_marke...

It's remarkable how commonly that link, between the USD and the vast US economy vs oil, is understood backwards. I think it's frequently misunderstood as an ideological result.

My other favorite: the failure to understand how a rising or falling dollar value is what has been triggering the commodity boom cycles, rather than primarily demand or magic animal spirits. The 1990s saw incredible global economic growth, and oil (gold etc) got cheaper and cheaper as the USD gained value toward the end of the decade. The global economy expanded dramatically from 1980 to 2000, yet the price of oil collapsed, despite demand going through the roof. It of course wasn't supply/demand that caused oil to go from $20 to $120 inflation adjusted, from ~1973 to 1980, it was a debased dollar; and it wasn't supply/demand that then caused oil to implode from $120 to under $20 (again, inflation adjusted) by 1999-2000. Similarly the GW Bush wars and economic policies + Fed policies heavily debased the USD, which sent oil, gold etc soaring and other nation GDPs soaring when priced in dollars. And then when a stronger dollar returned in 2014, post QE, oil plunged accordingly (and that was incorrectly blamed on oversupply of oil, when in fact oil had been in oversupply for years prior to it crashing). And again most recently, oil began climbing again, exactly when the dollar began falling from its highs (eg 100+ on the dollar index).

Modern 'economists' get these relationships entirely backwards. They also comically miss the way the US has been able to restrain Russia by using the USD to crash oil (their achilles heel), which was done in the Reagan years and toppled the USSR by bankrupting its finances (it wasn't military spending that did them in), and again immediately after the Ukraine invasion (brutally punishing their economy and making it difficult for Putin to continue to fund his foreign adventures).

> it wasn't supply/demand that then caused oil to implode from $120 to under $20

Yes, it was. On the supply side, we got fracking. On the demand side, we got Great Recession.

> restrain Russia by using the USD to crash oil

The recent oil crashes were prompted, again, by supply-side factors. Not government intervention in the U.S. dollar's exchange rate. If anything, American military interventions in Libya and Syria, combined with equivocation with respect to Iran, has introduced supply-side constraints on the market.

It's also remarkable how you can present only one side of the coin. Why aren't we all doing international trade in Rubles? Well, the military spending has at least something to do with that, the Soviet economy was not exactly a powerhouse compared to the US's.

And do you think the powers that be appreciated it when Saddam and Russia started to plan for an oil market denominated in EUR instead of USD?

That it exactly the sort of thing you might want to do as a sort of hedge against the restraint against Russia you mention. So while I agree with you 100% in the examples you describe, you should be aware that some talking about the "petrodollar" has this more nuanced understanding.

It will change the geo-political realities of world. For example, Petro-States will not wield the kind of influence they do now.

Energy changes Geo-politics of the world. Even Shale revolution in US changed many things around the world including ability to Russians to blackmail Europe on Natural Gas supplies.

The second and third order effects of "electrification" of transport are immense and most we really do not know as of now.

Transportation is still 92% petroleum


It will take some times to replace all existing vehicules by EV. So it might be a bit early to ask the question.

Prices are heavily linked to demand, further if demand starts to drop significantly some producers will try to dump as much as possible before their supply becomes 'worthless'. On the other hand many suppliers are rapidly running out of oil.

IMO, it's going to get interesting.

It is easier to convert an ICE to run on NG than electricity.

What clickbait nonsense and misleading nonsense.

Solar accounts for 1.3% of our electric production.


Saying that solar is the biggest source of "new" US Power is like if bill gates picked up a penny from the street and bloomberg saying picking pennies from the street is bill gates' biggest source of "new" wealth.

I understand being pro-solar, I am too. But why is HN and much of social media rife with lies and misrepresentations about solar?

Reading bloomberg and HN, you'd think solar was a significant contributor to energy and coal was dead. But that's just factually incorrect. And they been lying for years now. But most importantly, it's the scientific lies about energy production and solar that is bothersome. As if solar panels are the answer for coal.

Natural gas ( a fossil fuel ) is why coal has leveled off ( not died but leveled off ). Solar had nothing to do with it. And coal is still the biggest source of electricity around the world and the 2nd biggest source in the US.

Solar energy is going to be a minor part of the world's energy strategy unless a revolutionary breakthrough happens ( a significant breakthrough on par with cold fusion ). That's a basic fact of current state of physics.

So sick of misleading clickbait from media outlets like bloomberg.

Edit: If you want a total world view of energy sources.


I think the misunderstanding here is that you're talking about now: "solar is 1.3% of our electric production", "coal is still the biggest source", etc. This, I believe, is missing the point of the article.

The importance of noting that solar is the biggest source of new power is that is signifies the economics of power generation have changed. So sure, there's still lots of coal and there's still lots of natural gas. But if you're adding new capacity we're at the point where it makes more sense to build large solar arrays.

So in a sense, this article is noting the beginning of the end for coal and natural gas. The present percent contribution of solar or coal is not so important.

But this article doesn’t say anything about power generation. It only discusses nameplate capacity. Because the sun moves in the sky, and we have clouds, and night, and dust that sticks to panels, etc., the actual generated power is far less than the nameplate capacity.

For example, one set solar panels on my roof yielded 5.2MWh in 2017. Their nameplate capacity is 5kW, so the nameplate yearly generation is 5 * 24 * 365 = 44MWh, so my yield is about 12%. The other, smaller set (better panels, better site) yielded 4MWh or 15%. Actual generation is roughly an order of magnitude smaller than nameplate suggests.

And I live in California, where we don’t get a lot of rain - these numbers could be worse.

If you look at the power that will be actually generated by this new solar capacity referenced in the article, it’s far less than the power that will be generated by the other technologies, just because solar has such a bad yield.

So it’s not even true that incremental power generation is dominated by solar.

I love solar power, but I don’t think articles like this advance the cause, because they aren’t thinking critically enough. Lazy reporting does us no favors.

Don't let engineering numbers mislead you. The only number that really matters is cost per kwh at scale. Utilities are investing in this technology because that number has become favorable. We're starting to cross over from early, experimental deployments to bigger production. Utilities, which are very risk-averse, are becoming comfortable with solar.

12-15% capacity factor is comparable to a natural gas peaker plant.

Right, but the average capacity factor for natural gas plants in the US is about 56%.


The capacity factor for utility scale solar PV in the US was 27% in 2017:


The capacity factor gap with utility scale gas plants isn't as large as you might guess from looking at residential rooftop solar.

True. I guess the overall point I was trying to make is that the low capacity factor of solar strengthens the case for this beginning the real transition to renewables, since utilities are used to low capacity-factor assets (and are still choosing solar in spite of that drawback)

At some point within the next 3-5 years, storage + solar will exceed the capacity factor of natural gas without the fuel price volatility.

"New" is relative, hence the Bill Gates picking up the penny example. Sure, it's a true statement the second he picks up the penny, but it's not a meaningful one. In the case of the article, we are talking about a single quarter.

"New" is certainly meaningful. It gives an indication of where the market is going, investors want this information. It definitely meaningful to GE's natural gas turbine division, which has had big losses since demand has been less than they have forecasted

"Renewables boom is a bust for gas equipment suppliers like GE" https://www.utilitydive.com/news/renewables-boom-is-a-bust-f...

Sure, it has a meaning, but it does not have the meaning implied by the title, hence clickbait, which was paidleaf's point.

The title says "new" and the article discusses new generation. Paidleaf said the title makes an implication about overall generation, how does it do that?

It is meaningful here, though. If Gates started showing indications that he was primarily going to pick up pennies from now on, that'd have important implications and questions raised.

The article could have made a stronger case about the broader trends: although this is referring only to a single quarter's worth of new capacity in the US, this trend has been happening on a longer term in many other countries.

> The importance of noting that solar is the biggest source of new power is that is signifies the economics of power generation have changed

Not necessarily. Solar becoming the biggest source of new power doesn't mean that there aren't any new projects for gas plants etc., and if people are still building new non-solar plants then solar's economics aren't always better than non-solar sources, particularly since the economics of power generation are very much geographically dependent.

Solar becoming the biggest source of new power is an important milestone, but it's also a relatively meaningless statistic. More important would be measuring the rate of change of solar adoption in new power generation projects, if it's positive then either new solar projects are bigger than previous solar projects (for a constant rate of new projects year over year, meaning that solar is scaling well), or the number of solar projects is increasing (meaning that solar has become relevant in more and more areas), or both. Either way, if you're concerned about solar adoption, that's the real statistic to track.

I don't get why you think "new" is inaccurate or misleading. This shows where investments are being made now, which is highly relevant. For example, GE's natural gas turbine division has been taking a beating lately due to renewables, ie:

"Renewables boom is a bust for gas equipment suppliers like GE" https://www.utilitydive.com/news/renewables-boom-is-a-bust-f...

You make points about the overall contributions of fossil fuels to national generation capacity but that's not what the article is about.

Don't get me wrong, there are other reasons the headline and article are misleading. Comparing nameplate system capacity of electricity sources w/o taking into account the capacity factor is highly misleading. Also, discussion of whether solar projects were pushed up in an effort to beat the tariffs would have been appropriate.

"My deliberately incorrect reading of the headline is wrong!"

Speaking of misleading, your Bill Gates analogy is quite misleading too. These figures about new electric power are sums over a quarter-year, so small statistical fluctuations like finding a penny (or opening a single power plant) are smoothed out. To complete the analogy, if Bill Gates' biggest source of new wealth this quarter and last quarter were finding pennies, that would be quite a big deal, because over that period he would have to find hundreds of thousands of tons of them - more than doubling US zinc production in the process.

Or Microsoft's stock price took a dive? Analogies are fun!

>Solar energy is going to be a minor part of the world's energy strategy unless a revolutionary breakthrough happens

It doesn't really need a revolutionary breakthrough, just for solar and battery costs to drop 50% or so such that they become the cheapest option. A 50% drop will probably happen just from the existing trends continuing.

Indeed. I'd say we are in the midst of a "revolutionary breakthrough" right now. The cost-per-kwh of solar/wind electric power is now lower than fossil, and still dropping. Consumer options are growing on the low end, and experience with industrial-scale installations is growing on the high end.

There's another breakthrough I'm looking forward to, however - really good instantaneous pricing and metering. When prices are set on demand and fully automated, then energy caching hooked to the grid becomes viable. This creates a good arbitrage financial model. Store energy from solar/wind when the sun is shining and the wind is blowing, and sell that stored energy when demand is up and supply is down.

Once that's working, the whole grid gets remodeled. Then we can make a real dent in the coal-fired baseload plants and aging nuclear plants that desperately need retired. Economies of scale and experience will reduce production cost for whole suites of energy storage technologies (batteries, compression, thermal, flywheel, gravity, etc), even if they don't become more efficient.

That's an exciting future.

We don't even need the price drops you quoted. Installing new solar is already the cheaper than installing coal or gas in many locations. At this point we just need to continue to deploy. The panels themselves are already very cheap! Having out local governments and utilities simplify the permitting process would bring down some of the soft costs and make things even better though.

What this ignores is that the price for the end consumer is not directly related to the price of solar panel.

The combined cost of solar, batteries and grid changes, more advanced handling and software actually don't result in cheaper energy prices.

Compare California to places where they use gas.

Yeah that's kind of why we need another 50% or so to really be cheaper after you allow for the problems with peaker plants, upgrading the grid and so on.

I don't think it's misleading at all. More new solar power is being built than any other source. That's a good thing and if the trend continues, solar will eventually overtake other sources.

If, over the course of the past year, Bill Gates made more money from picking up pennies than any other source, that would be news too. It would indicate that his other sources of income have dried up.

You're right, the article doesn't say coal is dead.

It says coal is dying.

Social media is rife with what you call "lies and misrepresentations about solar" because a lot of people are very excited to see real progress being made in converting the electric grid from dirty, limited coal to clean, unlimited solar/wind. When I see numbers like this, I'm not thinking about "1.3%". I'm thinking about what this trend means for how it's going to look in 20-30 years.

Other countries have solved this today, even others 40 years ago (France).

But we today are exited about 'maybe in 20-30 years'.

Big projects take time. Decades. The financial models of power utilities are based on multi-decade deprecation of plants that cost hundreds of millions (or billions) to build.

Adding solar/wind to add new capacity or replace retiring capacity is a different and much better financial model than replacing equipment that is not yet fully deprecated. But it's a much slower one. Be patient.

I understand why it is not happening.

But my point is that there seem to be more interest in solar and wind as ends in themselves rather then actual discussion of how to de-carbonize the economy.

That just seems like an unnecessarily negative perspective to me. The point is, we are starting to de-carbonize the economy. And it's being done because market forces are a tailwind, not a headwind.

All the moral imperatives in the world won't be as effective as a profit motive. I'm not one of those silly people who finds a moral valence in market forces, but I'm always happy when I see them working in my favor, because they're as relentless as erosion.

The problem is that the energy market is far to regulated to make it clear what are 'market forces' and what are not.

Wind and Solar farms get production tax credits while producing useless energy (or even energy that they have to pay to dispatch). That gives them a significant advantage to something like a nuclear plant.

I am happy that these things are cheaper but I would also say that if we are seriously talking about this from a environmental perspective and how they will solve the problems we have.

If this is true then this endless talk about solar/wind is harmful to the larger debate. The ends and the means seem to be backwards knowdays.

Is it, though? Solar/wind is the path to getting us off fossil fuel. (I'm ignoring nuclear, as it has not managed to take off outside France, and the pro-nuclear contingent is so full of wishful thinking and conspiracy theory that it's like nails on chalkboard to listen to them.)

Early production credits are just priming the pump for solar/wind, helping the lead time to get the market going. They're neither necessary nor viable in the long run. Nor are they reliable for long-term utility commitments - all it takes is a political wave of fossil industry shills and anti-environmentalist fanatics (like, say, Washington today), and all those long-fought subsidies and credits can disappear. Worse, such subsidy/credit models are notoriously difficult to target effectively - the Obama administration, for all their good intentions, completely botched it.

But regulation and subsidy are, themselves, just market forces, more conditions to set supply and demand and therefore price. I don't believe in bullshit fantasies about frictionless free markets, and to the extent that they do exist, they just drive profit margins to near-zero (see the margins for supermarkets or gas stations). Show a real capitalist a regulation, and she'll see a market opportunity to be exploited. That's how capitalism actually works outside the whiteboard.

The killer for nuclear isn't regulation or subsidy - it's latency. Nuclear plants are extremely expensive to build (not because of regulation, but because of complexity and scale). This introduces much greater long-term financial risk, and the industry is full of failed boondoggles, not just in the US but everywhere.

Solar and wind, on the other hand, can be experimented with at a small scale, for very low cost, and lead times an order of magnitude less than nuclear. What we're seeing now is a scaling up of the initial experimentation, based on initial learnings.

In software terms, solar/wind allow for Lean product development. Nuclear is waterfall in its purest form. Which one gets us to viable products faster and more reliably?

Yet nuclear reactors used to be built much faster and still are in some places. There is clearly something going on.

If you call all these things 'market forces' then its meaningless to talk about market analysis. The winners are picked by the political process.

While solar and wind are easy to play around with its that does not get you to de-carbonised energy production.

We don't need to 'play around' we need to mass produce cheap carbon free electricity.

> why is HN and much of social media rife with lies and misrepresentations

Clickbait (economic or personal gain), or people perpetuating things that sound good, or things they want to believe. Emotion influences the perception of information, and information can influence emotion.

> Solar energy is going to be a minor part of the world's energy strategy unless a revolutionary breakthrough happens ( a significant breakthrough on par with cold fusion ). That's a basic fact of current state of physics

What physics based on what equipment? An array of photovoltaic cells may or may not be as efficient as the Ivanpah solar facility. Ivanpah produced 719,398 GWh/yr across a six square mile area. Theoretically we would only need to cover half of the Mojave with such sites to provide for the US's annual electricity need.

I don’t understand what you think is misleading here. It clearly says new power. That means it’s growing faster, not that it’s the biggest.

Your Bill Gates analogy doesn’t work. He’s still making fantastic amounts of money from his investments. If picking money up off the sidewalk was actually his biggest source of new wealth (which is an odd way of saying “income”) then that would be huge news.

> https://www.eia.gov/tools/faqs/faq.php?id=427

According to your source, 1.3% of utility production comes from solar. It provides an estimate for smaller residential rooftop installations which brings total current generation closer to 2%.

1.3% of generation today with a growth rate of roughly 30-40% per year. We are installing a ton of solar and are no longer building a significant number of new coal plants or nuclear plants. The article is pointing out an emerging trend. Before we know it, solar generation will account for 5% of electricity generation.

It's understandable coming from Bloomberg, their MO is to drive up clicks/views. Is it technically untrue? Well, in this advanced economy with projected flat electricity demand, there isn't much growth in demand for utilities, except for replacing retiring stock like coal, nuclear, etc. It's simply a massively large industrial base that fossil fuels are leveraging day-in day-out, powering the vast majority of vehicles, airplanes, cement, steel, aluminum, plastics, and on and on. Electricity is one component of our emissions base, yes, but it's something like 25-30% of our emissions. Solar/wind/other non-fossil-fuels will take 100 years to fill all of the use-cases seen with current fossil fuels, and even then they still may be used in some niche applications.

Sadly, readers don't want to face the uncomfortable facts: we are still a fossil fuel civilization and will be for decades.

When I flew to Munich a month ago for the D conference, I was impressed with all the rooftop solar and stretches of panels out in fields. I didn't do a count, but it looked like a good quarter to a third of homes had panels on the roof. Lots of random windmills here and there, too.

We need multifamily dwellings with the feature of no utilities bill because of local renewable energy on site. This would be awesome. The current vision of the future being sold to us seems to be single family homes with electric cars in the garage and solar cells on the roof. A horrible use of land and not really scalable to global population levels given the rise of cities.

> A horrible use of land and not really scalable to global population levels given the rise of cities.

Think outside the bubble, though.

There's unused land everywhere I look.

Why are we dissing zero-emission cars and self-production of home energy?

> Think outside the bubble, though.

A single family homes and car mindset is the bubble, though.

The US and much of the developed world has a problem of over-consumption, and shuffling to more power, some green, does not solve it. We need green energy, but if we want to reduce emissions we need to cut usage as well.

The amount of energy needed to cart single occupancy cars, on hour long commutes through sprawling suburbs of energy guzzling single family homes is horribly inefficient, no matter how it's powered. Nearly a third of the energy consumed in the US goes to transportation.

It makes no sense for multi-ton hunks of steel, plastic and lithium using tens of kW in energy to transport a 150 lbs person to work.

Denser cities and towns organized around public transit, cycling, and walking knocks out a multitude of problems.

1. Reduced distances between home, work, and shopping and allow for healthier, emission-free walking and cycling

2. Increased density reduces per capita cost on infrastructure and services and makes fast, frequent transit a realistic alternative

3. Denser multifamily development offers efficiencies in shared heating and cooling systems, a major consumer of energy

4. Cutting back on sprawl reduces the amount of land developed and contaminated by development, and frequently preserves productive agricultural land from being permanently developed

These aren't radical ideas. We have models throughout the world of greener, denser and healthier cities, mostly built before the auto.

Zero emission cars are great; how about an apartment building with an attached parking structure with free power coming straight from the roof?

How do you convince the middle class to accept this? It will be very difficult.

I think that much of the next generation middle class would gladly accept multi family dwellings if it means less driving, more walkable amenities, etc.

The problem is that there's not enough of that type of development out there, and land ib those types of settings is therefore very expensive.

Note that this 13% rise (1Q 2018 over 1Q 2017) is way below the 2010–2015 trend, even discounting 2016; the article shows US photovoltaic installations growing during that time by 55% per year. Trump is clearly having a huge negative impact. For better or worse, other parts of the world are not Trump-restricted, so PV is continuing its exponential trend there.

Paidleaf raises the reasonable question in https://news.ycombinator.com/item?id=17292776 as to why it matters when PV is only about 2% of US electric production (and thus about 0.6% of the total US energy maket). (The 1.3% number paidleaf gives is wrong; see https://www.washingtonpost.com/news/energy-environment/wp/20... for more details, for example.) The reason is that exponential-growth phenomena, like the adoption of any new technology, are tiny for a long time until suddenly they aren't. If we suppose a 30% annual growth rate and an asymptote somewhere well above current world marketed energy consumption, then three years before PV is the majority of world marketed energy consumption, it will be only 32%; three years before that, only 17%; three years before that, only 8%; three years before that, only 4%; three years before that, only 2%, like the US's current electricity market.

Of course, you could reasonably argue that an immense drop in PV's annual growth rate from 55% to 13%, or even the SEIA's projected 0% for all of 2018, indicates that we're already approaching the asymptote! And, if that's true, of course it's relatively inconsequential what happens with PV — it might eventually grow from 2% to 4% or 6% or something, but it'll never be a viable alternative to coal, and at worst it's a dangerous distraction from the necessary switchover to nuclear power.

But every indication on the fundamentals is that this isn't the case. We aren't running into limits on energy demand, on raw-materials availability, or on solar-resource availability. Photovoltaic energy remains cheaper than other sources, even in the US — that's why it's the majority of new generating capacity — and panels continue to drop in price, indicating ample raw materials and a smoothly running supply chain. The majority of even high-solar-resource land is still not shaded by solar panels.

(I'm not saying that shading the majority of high-solar-resource land with solar panels is a good thing or a bad thing. I'm saying that a scarcity of high-solar-resource land is not currently a factor that could cause solar adoption to level off.)

bcatanzaro comments that the capacity factor for the solar panels on his house is about 12% https://news.ycombinator.com/item?id=17293375 and so 55% of new installed capacity is actually much less when it comes to actual generation. But utility-scale solar's capacity factor is typically closer to 25% (I guess bcatanzaro didn't choose his house's siting and roof angle optimally for solar power generation, but utilities can; or maybe he just isn't dusting them) and the other new generation is almost entirely natural gas and wind, which generally also have very low capacity factors.

Every indication is that economic development is still solar, and economic development is still happening, despite the gradual growth of what appears to be WWIII. It's just that the US is missing out.

> The reason is that exponential-growth phenomena, like the adoption of any new technology, are tiny for a long time until suddenly they aren't.

My favorite exponential-thinking thought experiment is imagine a lake that just got an invasive plant injected into it. The plant has no predators and reproduces daily, so lets say it doubles every day. If we say the lake gets completely overrun in 30 days, what day is the lake half full?

There's a fancy formula for this, but you don't need it -- it doubles every day so it takes all the way until the 29th to be half full.

That's the key about renewables... they're not an extractive technology (where the more we extract only the more expensive stuff sticks around), they're a manufacturing technology (where the more we make, the better we get at making it and so it only becomes cheaper).

Slightly offtopic, but I do miss the blog http://www.theoildrum.com/ for discussing energy issues - it had the highest quality discussions of the subject I've ever found on the internet. Is there a current replacement?

I hope at least some of these panels are manufactured in the USA... Otherwise, it's another sector of the economy getting outsourced and impoverishing the north american economy in the long run.

If people on the left had the same obsession with nuclear that they have with solar climate change would be solved.

Honestly it seems to me like 'liberals' (lack of a better term) are closing their eyes on this. This eternal year on year constant media bombardment about the successes of solar energy and in the last couple of years batteries is totally out of proportion with any realistic view of the current energy system or any energy system that could realistically exist in the next 10 years.

What is ironic in this whole discussion is that while liberals can never stop pointing out how France is so much better in terms of health care system and child care and whatever else. The fact that France has abolished carbon in their energy production 40 years ago is somehow ignored.

A modern industrial nation abolished carbon in 1970 with technology that is essentially from the 50/60. This success if it had been done with solar and wind would fly on every flag of every news paper and environmentalist poster.

This could easily be repeated today in any other industrialized country. Switzerland and Sweden currently work like that (just with more water).

What is really sad is that the carbon industry has discovered this in the 70s already. Carbon based energy producers have long supported anti-nuclear campaigns because they know it will be incredibly difficult for solar/wind to remove goal/gas as baseload powers and thus they are only a very, very long term thread to them. While a nuclear plant could be plopped down right next to every coal plant.

I think to say this is a "liberal" problem is to miss the fact that more than a majority of independents, and nearly a majority of Republicans oppose nuclear energy http://news.gallup.com/poll/190064/first-time-majority-oppos... To put those numbers in context, it is important to contrast it with support for the competitors, solar and wind (http://www.pewinternet.org/2016/10/04/the-politics-of-climat... but it doesn't break it down by party affiliation), which have 80%+ support.

I guess 'liberal' is the wrong word. I mean the popular media and popular discourse in general.

One thing has to be admitted, the environmentalist campaign nuclear energy must have been one of the most successful ever.

The amount of Anti-Scientific nonsense that is populated and believed by large part of the population is even worse then climate change or anti-GMO. However Greenpeace and other organisation that spread these Anti-Scientific organisations never get called out on their Anti-Nuclear propaganda because to few people care.

Essentially a whole branch of technology has eliminated by these narrow minded propaganda campaigns.

Other technologies could have thrived even with this opposition but nuclear is to political and the regulation to restrictive. Its so sad given the mountain of different technologies we are missing out on.

Nuclear is too expensive (billions of dollars per facility) and takes too long to build (10+ years), full stop. This doesn't even address the waste storage issue.

France is going to need to address its expensive, aging nuclear generators at some point. By then, of course, solar will be much cheaper than it is today, and there will be more HVDC transmission lines shuttling power between EU countries.

Nuclear is expensive in the US NOW because of insane amounts of regulation, constant law-suits and the fact that almost no new plants were build in the last 30 years.

Back before 3 Mile Island a nuclear plant was cheaper to produce then coal plant and had 2x the amount of live time with 10x cheaper fuel cost.

> France is going to need to address its expensive, aging nuclear generators at some point. By then, of course, solar will be much cheaper than it is today, and there will be more HVDC transmission lines shuttling power between EU countries.

France has the lowest electricity prices in Europe had them for quite a long time.

Nuclear is not expensive if you are building many of them, just like everything else.

One of the major problems with nuclear is that it is extremely national and most western nations don't have a demand for scale.

Back before 3 Mile Island

And then the 3 Mile Island disaster happened, and people tighten up regulations. That seems reasonable, no?

And Chernobyl. And Fukishima. Yes, nuclear disasters are rare, but when they do happen it takes extraordinary resources and time to remediate. Solar panels and batteries fail more gracefully.

Even that is overestimated.

Solar panels create mountains of waste, much of it being disassembled by kids in Africa who then get sick because of it.

Solar production facility can have chemical spills that are comparable to Three Mile Island in terms of effect.

Compare a nuclear civilization against a solar one and you will see that waste management of nuclear is far smaller.

Overall less then 2000 people died from these 3 disasters and essentially all of them because Chernobyl that uses a technology that we don't use in the west.

"Solar panels create mountains of waste, much of it being disassembled by kids in Africa who then get sick because of it."

Source? Electronics associated with solar panels might be an issue but modern inverters use less and less toxic elements (and are definitely RoHS compliant).

As for solar panels themselves, 99% by weight consists of glass (75%), plastics (10%), aluminum (9%) and silicon (5%) which can all be recycled.

I don't want solar to be the enemy. I just wanted to point out that its not quite so simple and that 'nuclear waste' is a far smaller problem then people realize.


Your reaction underscores the point that fears over nuclear power stem from widespread lack of education about nuclear energy as a whole. TMI was blown ridiculously out of proportion. A pressure valve got stuck resulting in a leak that was... wait for it... undetectable compared to background! Does shutting down a comparatively clean energy supply and driving an entire industry effectively out of business seem like a rational response to that?

Nobody in the nuclear industry was against some of the changes and more regulation.

The problem was that it turned into an absolute shitshow from a regulator perspective, process perspective, utility regulation perspective, political perspective and so on that. There were 100 plants in construction and all of them go canceled.

Large part of the nuclear industry when bankrupt and so on.

Since then regulation has prevented most progress on any of these issues. Nuclear startups all move to Canada, China or some other place because the US is simply not good place to work on nuclear.

The actual changes that would have been needed after Three Mile Island were mostly about the human to reactor interaction, ie. UI. Why that should destroy and industry make the cost 10x higher is quite hard to understand.

Umm...so who pays for cleanups and other damage in situations like Fukushima? Perhaps you should read this article before trying to promote something like this: https://www.scientificamerican.com/article/clearing-the-radi...

Nuclear plants may be expensive, but it's still the cheapest energy out there. It's kind of misleading when you talk only about the cost of nuclear facilities and not about the efficiency in the long run. There are lots of legit arguments against nuclear energy, yet you chose to complain about the cost of the facilities or the time it takes to build them. Your only real point was the waste storage issue (the aging reactors problem is essentially the same as the cost problem).

An airplane is also more expensive than a bike, and takes longer to assemble. This doesn't mean it has no use.

Nuclear power is one of the most expensive sources around.

See Lazard's 2017 numbers[1], which put the low price of nuclear power at $112/MWh. Only rooftop residential is more expensive ($187/MWh), Diesel generators ($197/MWh) and gas peaking ($156/MWh) are more expensive.

Utility solar is as low as $43/MWh, wind is $30/MWh

If you think there is going to be some magical world where people get rid of safety regulations for nuclear power, after the Fukushima nuclear accident I think you need to reconsider. Nuclear power advocate's excuse for that one was that they had lax safety standards, so dropping them further seems unlikely.

[1] https://en.wikipedia.org/wiki/Cost_of_electricity_by_source#...

In 80s 100 reactors were in planning and the price was lower then a coal plant.

Many of those are still running and they are safe.

Understanding why the cost has gone up 10x can not be explained simply by 'slightly higher security regulation'.

Something else went wrong in the whole process and the regulated utility market.

> If you think there is going to be some magical world where people get rid of safety regulations for nuclear power, after the Fukushima nuclear accident I think you need to reconsider.

Nobody wants to 'get rid' of safety regulation. However talk to any expert on nuclear energy and you will see that the current way to to regulation is highly ineffective and damages the industry.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact