US solar power generation for 2014-17

[nikkkom]

Solar generation results for US are out for 2017.
Year over year:

http://www.eia.gov/electricity/annual/html/epa_01_02.html
https://www.eia.gov/electricity/annual/pdf/epa.pdf

US: Net Generation (Thousand Megawatthours) 2014
Solar: 17691 (that's 0.43%)
All sources: 4093606

US: Net Generation (Thousand Megawatthours) 2015
All sources: 4077601
Solar utility scale: 24893 0.61%
Solar all (estimated): 39032 0.95%

US: Net Generation (Thousand Megawatthours) 2016
All sources: 4076627
Solar utility scale: 36054 0.88%
Solar all (estimated): 54866 1.34%

US: Net Generation (Thousand Megawatthours) 2017
All sources: 4034268
Solar utility scale: 53286 1.32%
Solar all (estimated): 77276 1.92%

 

[anorlunda]

Thanks for sharing. The public confuses the soaring growth rate of solar (and wind) with the very low starting point (sub 1%). The public also finds it hard to comprehend an industry that requires 20-40 years to make big changes.

A very interesting development is that utility-scale contracts for energy futures have been bid and won by solar or wind including storage.

 

[russ_watters]

Yes, thanks for pointing to this. This is an issue I've been tracking and I have a spreadsheet and some analysis I'll share later.

We've previously discussed growth rates and scalability, and I have commented largely on two benchmarks:
1% (passed in 2016) was my threshold for considering solar power "relevant", meaning prior to that it was just too low to affect the grid or make realistic predictions about its growth. Opening one plant would skew a year's growth rate. We've passed that and growth rates have settled down.

10% is suspected to be the limit of practical exploitation based on Germany's model. We're about 5 years from that based on the current growth rate trend (about 40% now, and dropping pretty much every year).

Obviously the industry climate has a big impact on what happens next.

 

[nikkkom]

Solar utility scale generation
growth 2014 to 2015: 24893 / 17691 = 1.407, or ~41%
growth 2015 to 2016: 36054 / 24893 = 1.448, or ~45%
growth 2016 to 2017: 53286 / 36054 = 1.477, or ~48%

 

[mfb]

10% is suspected to be the limit of practical exploitation based on Germany's model.

The US has places with more sunlight, but it doesn't feature the ridiculously high subsidies Germany has.
The US needs more electricity in the summer, that works nicely with solar power.

 

[phyzguy]

Storage is the key. Sunny places like Hawaii and California are already facing problems where generation exceeds demand on sunny summer days. Until we have efficient utility scale storage, it will be hard to have solar power go beyond 10% or so.

 

[Tom.G]

from: https://www.nytimes.com/2018/05/09/business/energy-environment/california-solar-power.html

California law requires at least 50 percent of the state’s electricity to come from noncarbon-producing sources by 2030. Solar power has increasingly become a driver in the growth of the state’s alternative energy production.

At the end of 2017... Solar power provides almost 16 percent of the state’s electricity...

 

[OmCheeto]

... problems where generation exceeds demand ...

What a horrible problem. This has also been an intermittent "problem" with wind farms, also.

Storage is the key.

I agree with this.

Until we have efficient utility scale storage, it will be hard to have solar power go beyond 10% or so.

I disagree that "utility" scale storage is the only solution.
Being a northerner, I could really use some "thermal" storage, similar to the Drake Landing type of system.
My annual heating requirements are 3 times that of my base load.
Reducing my heating bills to zero, I could get by easily with a single Tesla Powerwall.

My base load: 9.0 kwh/day
Powerwall: 13.5 kwh​

http://www.eia.gov/electricity/annual/html/epa_01_02.html

That sure is a fun page to ogle. I've created countless interesting graphs from it.

Quite the impressive reduction in SO₂ over the last 10 years.

I'm guessing that's mostly because of a shift from coal to natural gas.
It will be interesting to see what it is in another 10 years, as solar increases.

And imagine what the following graph would look like, if everyone followed the Drake Landing example:

Per wiki; "In the USA, HVAC systems account for nearly half of the energy used in residential buildings."
[edited for brevity]

As I've mentioned before, I will do what I can, when I can afford it*. Commercial and Industrial customers can figure it out for themselves.

----
* Coincidentally, on the day California senate bill 100 passed(Aug 28th), a rep from from my electric company showed up on my doorstep and talked me into signing up for 100% renewable energy. According to them, it will cost me an extra 5.5%. I still have no idea how they are going to provide me with 100% renewable energy, when such a thing doesn't exist, as far as I can tell, but, I did it anyways.

Also, by this day in 2021, I should be able to afford to both cut my neighbors trees down, and install a thermal sump.
Amazing what you can do, when the house gets paid off.

 

[phyzguy]

from: https://www.nytimes.com/2018/05/09/business/energy-environment/california-solar-power.html

California law requires at least 50 percent of the state’s electricity to come from noncarbon-producing sources by 2030. Solar power has increasingly become a driver in the growth of the state’s alternative energy production.

At the end of 2017... Solar power provides almost 16 percent of the state’s electricity...

This site, from ca.gov, says 10.2% in 2017, not 16%.

 

[phyzguy]

What a horrible problem.

It is a problem. Without a way to store the energy being generated, what is the utility supposed to do with it? Also there is a problem in that the grid, as currently designed, works well when power is flowing from the utility to homes and businesses, and can fail when power begins to flow back the other way.

I disagree that "utility" scale storage is the only solution.

By "utility scale", I just mean enough energy storage to store several days of demand, if it is cloudy or raining, etc. I don't mean it has to be at the utility. It could be many Tesla Powerwalls or some other form of local storage distributed across millions of users. But if you look at the numbers, it is a daunting challenge. California uses about 800 GWh of electricity per day. If we want to store 2 days of this usage, we need 1600 GWh of storage. A Tesla Powerwall stores 14 kWh. So we would need 114 million of them. This is a tall order.

 

[Tom.G]

This site, from ca.gov, says 10.2% in 2017, not 16%.

Ahh, well, it WAS from a newspaper; ( must have been 'Good Enough.')
Thanks for the correction! Appreciated.

1600 GWh of storage. A Tesla Powerwall stores 14 kWh. So we would need 114 million of them. This is a tall order.

Sure is tall. There are only:
California
Total housing units: 13,911,737
Number of Companies: 3,548,449
(and another order-of-magnitude creeps in )
That's 8+ Powerwalls each. Ouch.

source: https://factfinder.census.gov/faces/nav/jsf/pages/community_facts.xhtml?src=bkmk

 

[nikkkom]

By "utility scale", I just mean enough energy storage to store several days of demand, if it is cloudy or raining, etc. I don't mean it has to be at the utility. It could be many Tesla Powerwalls or some other form of local storage distributed across millions of users. But if you look at the numbers, it is a daunting challenge. California uses about 800 GWh of electricity per day. If we want to store 2 days of this usage, we need 1600 GWh of storage. A Tesla Powerwall stores 14 kWh. So we would need 114 million of them. This is a tall order.

Tesla Powerpack is more appropriate product to consider for mass storage. 210 kWh per fridge-sized box. You'd need "only" 7.6 millions of those.

 

[mfb]

Tesla Powerpack is more appropriate product to consider for mass storage. 210 kWh per fridge-sized box. You'd need "only" 7.6 millions of those.

At $80,000 per piece that would cost $600B, or $15,000 per capita - if anyone could build so many batteries.
Tesla's Gigafactory 1 is expected to reach 150 GWh/year production. We need 10 years of full production to be able to store 2 days worth of California's electricity. If the batteries last 10 years the factory has to run continuously just to keep up with aging batteries.

 

[OmCheeto]

It is a problem. Without a way to store the energy being generated, what is the utility supposed to do with it?

I'm not a utility, but when I don't need energy from my solar panels, I just unplug them. I do understand that there are "contractual" problems, with the "utilities".

Also there is a problem in that the grid, as currently designed, works well when power is flowing from the utility to homes and businesses, and can fail when power begins to flow back the other way.

Sounds like a problem to be solved.

By "utility scale", I just mean enough energy storage to store several days of demand, if it is cloudy or raining, etc. I don't mean it has to be at the utility. It could be many Tesla Powerwalls or some other form of local storage distributed across millions of users. But if you look at the numbers, it is a daunting challenge. California uses about 800 GWh of electricity per day. If we want to store 2 days of this usage, we need 1600 GWh of storage. A Tesla Powerwall stores 14 kWh. So we would need 114 million of them. This is a tall order.

Huge numbers divided by lesser huge numbers = manageable numbers, IMHO.

Fortunately for me, I have a mole(my sister) who lives in a region(San Diego metro area) that recently(last year) provided me with her utility bill history.

Her average "electrical" consumption (16 kwh/day) is quite close to your 19 kwh/day, so I'm going to call her a good source of empirical data.

And as far as I can tell, more than half (22,000,000) of Californians live the the Los Angeles and San Diego statistical areas.

Now, I may not be a genius, but if it's hot enough that you have run your a/c, then the sun is probably shining. Hence, no need for "electrical" storage. And given that "solar thermal" is 10 times as efficient as "solar pv", I'm going to guess that heating is 10 times less a problem, area and cost wise.

At $80,000 per piece that would cost $600B, or $15,000 per capita

This kind of brings my sister's next door neighbor into the equation.
He paid umpteen thousands[aka bazilions!] of dollars to install a 5 kw solar pv array, 20 years ago. Why should he have to subsidize those who couldn't afford it?

ps. My sister installed insulation in her attic this year. I'll ask her next year, in March, what happened to her utility bills.
Yay science!

 

[mfb]

And given that "solar thermal" is 10 times as efficient as "solar pv", I'm going to guess that heating is 10 times less a problem, area and cost wise.

What does "10 times as efficient" mean? The sunlight->useful energy conversion can't be 10 times better, solar PV has more than 10%.

He paid umpteen thousands[aka bazilions!] of dollars to install a 5 kw solar pv array, 20 years ago. Why should he have to subsidize those who couldn't afford it?

It is probably the opposite: Your neighbor got thousands of dollars as subsidy.

 

[CWatters]

I'm not a utility, but when I don't need energy from my solar panels, I just unplug them. I do understand that there are "contractual" problems, with the "utilities".

People get unhappy paying the utilities not to generate. If you don't pay them why should they build the plant?

It's not a problem if they only have to curtail generation infrequently but the problem with solar and wind is the variability. You have to install lot more peak capacity than you need on average.

 

[Rive]

Storage is the key.

I too used to think that, but by now... Well, I think the actual key is the equipment price falling so low that PV could be curtailed (remotely) without suffering noticeably loss (and gaining by the ability to curtail).

All that storage business - the numbers are just horrible.

 

[phyzguy]

I too used to think that, but by now... Well, I think the actual key is the equipment price falling so low that PV could be curtailed (remotely) without suffering noticeably loss (and gaining by the ability to curtail).

All that storage business - the numbers are just horrible.

If you rely on PV without storage, where will your power come from at night?

 

[Rive]

If you rely on PV

I don't want to rely on PV: I just want the mid-day consumption peak being mostly transferred to PV.
That's big enough goal for my lifetime

 

[phyzguy]

I'm not a utility, but when I don't need energy from my solar panels, I just unplug them. I do understand that there are "contractual" problems, with the "utilities".

I have solar panels on my house also. Averaged over the year, I generate more electrical energy than I use. But I don't have any energy storage, so I get power from the utility at night and in the winter. How do you deal with this? Do you have energy storage in your home, and if so, what kind?

 

[phyzguy]

I don't want to rely on PV: I just want the mid-day consumption peak being mostly transferred to PV.
That's big enough goal for my lifetime

That's a good goal, but I don't think it transitions us off fossil fuels fast enough to meet climate change goals.

 

[litup]

What about the economics of non-electrical power storage like deep underground caverns pumped with air and converted with turbines or some such? Of course that would not be a solution for home owners but utilities could do it. The question is the cost compared to Tesla batteries?

 

[OmCheeto]

What does "10 times as efficient" mean? The sunlight->useful energy conversion can't be 10 times better, solar PV has more than 10%.

I rounded.

It is probably the opposite: Your neighbor got thousands of dollars as subsidy.

My sister's neighbor. Solar pv would not work at my house.
And anyways, solar pv is so cheap now, I don't see any reason to continue the subsidies.

People get unhappy paying the utilities not to generate.

This is a media problem, IMHO. From my recollection, the amount of time wind utilities are required to shut down, is trivial. Haven't really looked into solar, as fall/winter/spring solar potential is kind of nil where I live.

If you don't pay them why should they build the plant?

Why I said; "I understand the 'contractual' obligations."

It's not a problem if they only have to curtail generation infrequently but the problem with solar and wind is the variability. You have to install lot more peak capacity than you need on average.

I agree.

I have solar panels on my house also. Averaged over the year, I generate more electrical energy than I use. But I don't have any energy storage, so I get power from the utility at night and in the winter. How do you deal with this?

As I alluded to above, I don't. My panels are just a hobby.

Do you have energy storage in your home, and if so, what kind?

If I did have access to southern California levels of sunshine, I'm sure I'd figure it out. But since I don't, I haven't.
We have stretches of MONTHS without sunshine.
Hence, why that solar thermal installation in Canada, so intrigues me.
My "off base" requirements mostly involve the "heating" portion problem of "HVAC".

Refrigeration will require a new refrigerator. Mine is actually quite old, and the nameplate data on new ones indicates and I could cut my base load down to 2 kwh/day with a very small investment.

ps. Got rid of my "tropical" fish.
pps. Hey! That reminds me. I never solved how much energy my back porch light was costing me:

3958 hours ≈ 165 days
5.62 kwh
yields 0.034 kwh/day. (2.8 watts continuous)​

Quite a bit higher than my estimated 0.0083 kwh/day. (0.7 watts continuous)
But still, quite reasonable.
I love my Kill-A-Watt [hour] meter.

 

[mfb]

And anyways, solar pv is so cheap now, I don't see any reason to continue the subsidies.

As an example, Germany subsidizes new installations with 12 cent/kWh. That is about three times the typical market price of electricity. In addition you are guaranteed that you can sell it - even if there is more production than demand on a sunny summer day. The government will still happily give you 12 cent/kWh for pushing electricity into a grid where the prices get negative, i.e. switching off your solar panels would actually be better for the grid. In addition you get some tax benefits and potentially local incentives for PV installations.
Sound great, right? Everyone must install PV like crazy! Well, turns out new installations have been dropping a lot over the recent years. New installations peaked in 2012 (when the subsidies were much higher than today), we are now below 20% of this peak.

The US has areas with much more sunshine and often area itself is more easily available, but the modules and their installation still cost a lot of money. In exchange you get something that is available most days, but not all of them. You still need conventional generation as backup. If this conventional generation is based on fossil fuels then solar power saves CO2 emissions - that is great - but it doesn't save the money needed to install and run these power plants. Just dividing the cost of solar power by its power output is misleading, it hides the costs of keeping a backup in case the Sun doesn't shine. The owner of the installation might make money, but overall it makes electricity more expensive because the operators of the power plants will charge more in a scenario with a lot of solar power.

 

[tech99]

There seems to be a growing public view in Europe that we are heading for an extinction if fossil burning continues. This thread does not give much confidence! The 40 year time scales alone seem to be cause for great concern as the window is closing.

 

[nikkkom]

As an example, Germany subsidizes new installations with 12 cent/kWh. That is about three times the typical market price of electricity. In addition you are guaranteed that you can sell it - even if there is more production than demand on a sunny summer day. The government will still happily give you 12 cent/kWh for pushing electricity into a grid where the prices get negative, i.e. switching off your solar panels would actually be better for the grid. In addition you get some tax benefits and potentially local incentives for PV installations.
Sound great, right? Everyone must install PV like crazy! Well, turns out new installations have been dropping a lot over the recent years. New installations peaked in 2012 (when the subsidies were much higher than today), we are now below 20% of this peak.

Germany is a very densely populated country. There are no big sunny deserted places. Also, 47+ degrees northern latitude... not great insolation (most of the CONUS territory is more southerly than that!).

 

[nikkkom]

What about underground caves used to store compressed air? Can that compete with Tesla batteries? Obviously not for homes but for utilities? Sorry, duplicated post.

It does not work that great. Air gets hotter as it is compressed. Then it cools down in the cavern. Then it gets cold when it's expanded through a turbine. IIRC sometimes up to partially becoming liquid :/

 

[nikkkom]

What about underground caves used to store compressed air? Can that compete with Tesla batteries? Obviously not for homes but for utilities? Sorry, duplicated post.

There are few other ideas (apart from ones in common use already, like hydro-storage).

(1) Use excess electrical energy to synthesize fuel from easily available source materials (such as air or water). Burn this fuel at night. (On Mars, probably use "air" as a source? CO2 -> O2 + CO). Does not seem to be economically possible on Earth: our air is not CO2, water electrolysis is probably way too inefficient (+ hydrogen storage is not easy), etc.

(2) Use excess electrical energy to synthesize higher-energy compounds from suitable stored source compounds. React these compounds (back into the original compound) at night. Rinse, repeat.
Usually, this takes a form of electrolytic battery where electrolyte amount is not fixed and not all contained in the battery, but is pumped to/from tanks of possibly much larger volume. https://en.wikipedia.org/wiki/Flow_battery, https://en.wikipedia.org/wiki/Vanadium_redox_battery.

 

[anorlunda]

How much wind/solar can the grid accept? Here are a couple of studies.

https://www.nrel.gov/docs/fy16osti/64472.pdf
https://www.pjm.com/~/media/committees-groups/subcommittees/irs/postings/pris-executive-summary.ashx

I trust private studies more than government studies because they are performed by reliability engineers. Being right on the forecasts is the reliability engineer's only agenda. They are very conservative professional planners. It would be very difficult to think of factors that they did not account for. They have all the data, all the confidential plans, powerful simulation software, and lots of time available to do their studies. But in this case, the government and private studies agree. [Edit: unfinished sentence fixed.]

What is the conclusion? Basically, that the US Northeast states could tolerate up to 30% solar+wind using existing transmission facilities, without major changes, and without compromising reliability. The studies do not assume much more storage than already exists. Additional energy storage developments would allow even more wind+solar.

30% is vastly more than today's wind+solar capacity, and vastly less than the 100% that renewable proponents suggest.

Participants in the debate should understand that local realities dominate over general principles. That makes public debates difficult because the public does have access to the local details since 9/11. It is also tedious for the public. Thost two studies linked here add up to 263 pages of mind-numbing detail.

But some local details that I know from my history in that industry.

• The US NE states have significant fast-response capacity in the form of hydro, pumped hydro, fast-start gas turbine and demand response resources already existing. Fast-response is the key to co-existence with solar+wind. Those resources contribute credibility to the 30% claim.
• There is plenty of wind power in the Midwest. Investors there wish to sell that energy in the Northeast where prices are higher. However, the existing transmission capacity is not adequate to ship all that power. At the same time, the Northeast states are happier buying electric energy from Ontario and Quebec. The capacity, the speed of response, the financial details, and the transmission capacity (both existing and new), of the Canadian power are more attractive than Midwest power. That US/Canada competition makes it a politically volatile topic with much lobbying of Congress. Midwest investors want the NE to be ordered to buy Midwest power, and for the NE to pay the costs of building new transmission.

 

[phyzguy]

@anorlunda - thanks for posting this. Wind+solar helps, because the combination is less variable than either one alone. Solar always switches off at night, whereas the wind still blows at night, for example.

Did you mean, "...the public does not have access to the local details since 9/11..."

 

[anorlunda]

Solar always switches off at night, whereas the wind still blows at night, for example.

That too is local. In the West, wind tends to blow at night. In the East it blows mostly daytimes.

Did you mean, "...the public does not have access to the local details since 9/11..."

Yes. Someone must have pointed out that those details could be used to plan a terrorist attack on the grid. Prior to that, most of the data was public.

I recall from my own days of transmission planning, of noting that a specific transformer in a specific substation, was especially vital to the grid in that unnamed country. In today's hostile world, it doesn't make sense to make such info public.

 

[mfb]

It does not work that great. Air gets hotter as it is compressed. Then it cools down in the cavern. Then it gets cold when it's expanded through a turbine. IIRC sometimes up to partially becoming liquid :/

Combine it with an air liquefaction plant?
There are a couple of creative options to make demand follow the production, at least partially. Some industrial processes are very energy-demanding, if you increase the capacity (to avoid the need to run 24/7) you can run them only when there is excess energy. That has higher installation costs but saves some money from electricity prices. Hydrogen production from electrolysis is one example.

 

[Rive]

What about underground caves used to store compressed air?

There are many really creative ideas. One of the really 'inspiring' ones is officially called 'hydraulic rock storage'. We often got spammed with that thing on a forum where I was familiar before. When the spamming got boring we started to call the pumped up state of the thing on a name what would make SpaceX's 'BFR' blush.
Spamming stopped.

Right now the financials are short even for the classic, cheap pumped hydro. Without drastic changes that price level makes those further ideas just empty dreams.

 

[nikkkom]

Combine it with an air liquefaction plant?

Well, the point was to recover stored energy. With unintended air liquefaction, that goal fails...

IIRC what they do currently is they inject and burn a bit of natural gas in the compressed air just before it enters the turbine.

 

[mfb]

With unintended air liquefaction, that goal fails...

You still get some energy back, but now you save a lot of the energy needed to liquefy the air. Liquid air (or liquid nitrogen and oxygen) is needed anyway. If you can use the plant to recover some energy on demand: Even better.

Independent of that: There are also approaches to store some of the produced heat from compression in liquids, as molten salt or whatever, so you can recover it later when the air is expanded again.

 

[nikkkom]

You still get some energy back, but now you save a lot of the energy needed to liquefy the air.

What would be the cost of transporting it to the users? LOX is much cheaper than petrol, thus it is usually it is produced right at the use site. Needs only air and electricity.

Independent of that: There are also approaches to store some of the produced heat from compression in liquids, as molten salt or whatever, so you can recover it later when the air is expanded again.

Yes. But ~30% conversion efficiency is a problem.

 

[anorlunda]

People get unhappy paying the utilities not to generate. If you don't pay them why should they build the plant?

The net metering scheme is unstable. It would lead to a death spiral for utilities. So eventually the model will have to shift toward utilities providing backup service to homeowners who are mostly self-sufficient. Buying backup is analogous to paying someone to not generate. How much should backup cost? That is a classical supply and demand problem.

Also remember that roughly half the people in the USA live in multifamily dwellings (or high rise apartments). They are not able to own personal wind or solar, so they remain dependent on the power grid sending power from remote places. Discussions like this one sometimes forget that not everyone owns a house.

have solar panels on my house also. Averaged over the year, I generate more electrical energy than I use. But I don't have any energy storage, so I get power from the utility at night and in the winter.

Have you thought that you are effectively using the grid as a battery? With net metering, they are providing you this energy storage service free to you, but paid for by other people. Why should you spend your money to buy a Tesla Powerwall if you can get someone else to pay the bill? That is why I say that net metering is an unstable business model in the long run.

I have a friend in Vermont with lots of solar panels. His utility found it troublesome for him to have excess generation. Not only financially, but also the problem of having power flowing the wrong direction in the distribution wires. (see # 10) The solution? The utility bought two Tesla Powerwalls for his house. It sounds practical, but think of the instability of the system that makes some people spend their own money, while other people get the same stuff free.

There are few other ideas (apart from ones in common use already, like hydro-storage).

All ideas are welcome, but remember that energy infrastructure is still financed by private investors. They have to be convinced in the safety of their investments and a ROI high enough to invest in energy rather than buy Apple shares. Prior to 1973, rich utilities used to finance a few high-tech speculative pilot projects every year to push the state of the art. After 1973, that is no longer allowed and private investors have to fill that role. As far as I know, no private investor in any country is forced to put their money into energy infrastructure.

 

[phyzguy]

Have you thought that you are effectively using the grid as a battery? With net metering, they are providing you this energy storage service free to you, but paid for by other people. Why should you spend your money to buy a Tesla Powerwall if you can get someone else to pay the bill? That is why I say that net metering is an unstable business model in the long run.

I agree. The current system only works when solar is a small percentage. It breaks down if solar becomes a large percentage. That's why I said (post #6) that we need large scale energy storage if you want solar to be the main source.

 

[CWatters]

Here in the UK the government has a scheme to help keep some power stations available as backup for renewable. The EU courts have just recently ruled it illegal.

Sent from my Hudl 2 using Physics Forums mobile app

 

[russ_watters]

[for when you return]

Solar utility scale generation
growth 2014 to 2015: 24893 / 17691 = 1.407, or ~41%
growth 2015 to 2016: 36054 / 24893 = 1.448, or ~45%
growth 2016 to 2017: 53286 / 36054 = 1.477, or ~48%

While I didn't specify which stat I was referring to, I was talking about total solar generation, as indicated to my reference to the 1% threshold passed in 2016. I think the total solar is the more relevant stat (you did, afterall, include both in your OP) as the utility and non-utility solar are part of the same market and competing with and otherwise affecting each other. I think this is why the total solar data is much smoother. Also, I look at the monthly data, which oddly is a touch different from the annual data, but allows a look ahead into 2018:
https://www.eia.gov/electricity/data.php#generation

The past few years of total solar growth have been:
2013-13: 133%
2013-14: 133%
2014-15: 120%
2015-16: 47%
2016-17: 43%
2017-18: 31% (est; likely to be lower)

Utility growth had been steadier, but lower than small scale PV growth until 2017, but small scale PV growth is dropping faster. You can see how the various growth rates are converging here:

Note, there is a buffer on the total solar and total utility solar (which I don't cite) in that solar thermal is in them. Solar thermal has apparently fizzled-out and isn't really growing.

 

[russ_watters]

The US has places with more sunlight, but it doesn't feature the ridiculously high subsidies Germany has.
The US needs more electricity in the summer, that works nicely with solar power.

Yes, there is a push-pull of different factors and we'll just have to wait and see how it shakes out. My prediction is that based on talk of just how absurdly high German subsidies have been, solar is way over-sold in Germany. The US doesn't have anywhere near as high subsidies, so I predict that downward push on our adoption will make a bigger difference than our stronger sun. But we'll see.

 

[russ_watters]

* Coincidentally, on the day California senate bill 100 passed(Aug 28th), a rep from from my electric company showed up on my doorstep and talked me into signing up for 100% renewable energy. According to them, it will cost me an extra 5.5%. I still have no idea how they are going to provide me with 100% renewable energy, when such a thing doesn't exist, as far as I can tell, but, I did it anyways.

I took a "job" interview with Green Mountain Energy a while back, and here's how it works: you can bill the generation separately from any source you want, as long as you keep track of the allocation. So it's just a spreadsheet exercise to say that all of your electrons came from 100% sources.

The real question is: does that help anything? It depends. Green Mountain Energy had plans where they were buying electricity from 50 year old, bought-and-paid-for hydro plants and re-selling it at a tidy profit without providing any additional environmental benefit. But they also had plans for wind and solar, which at the low adoption rates of 15 years ago meant that they had to build wind and solar plants in order to sell those rate plans. That did help -- but it was much more expensive. So read your fine print!

 

[anorlunda]

Here in the UK the government has a scheme to help keep some power stations available as backup for renewable. The EU courts have just recently ruled it illegal.

Until Brexit, correct? Don't forget that you have to preserve the power grid as well as some power plants.

s the utility and non-utility solar are part of the same market and competing with and otherwise affecting each other.

That's very true. Good data Russ. The stat I wish they would start publishing is the renewable+storage package deal utility contracts honored. It is a bit different than installed capacity because someone can bundle solar from here, wind from there, and storage at a third location to offer a package deal. During the validity of the contract, the physical assets used to fulfill the obligation can change. Such contracts are very attractive to utilities.

BTW, can't find the link, but I saw on reddit that US prices for PV panels imported from China, fell 20% this year despite a new 30% tariff.

 

[russ_watters]

There seems to be a growing public view in Europe that we are heading for an extinction if fossil burning continues. This thread does not give much confidence! The 40 year time scales alone seem to be cause for great concern as the window is closing.

Don't sweat it. There is a straightforward one word solution to our clean electricity needs for the next few centuries that we can do if we choose to*. We just need to wait a few more years until the solar fad wears off. The false hope of solar is the real damage it is doing by shifting focus away from solutions that can actually provide all our electricity needs.*Nuclear.

 

[davenn]

Don't sweat it. There is a straightforward one word solution to our clean electricity needs for the next few centuries that we can do if we choose to*. We just need to wait a few more years until the solar fad wears off. The false hope of solar is the real damage it is doing by shifting focus away from solutions that can actually provide all our electricity needs.

*Nuclear.

Even our Australian govt has started to seriously consider the nuclear solution
The problem is they are still too scared to implement it because it's likely to cost them the election win because of a people backlash.
So it continues to sit there in the background simmering away and the people continue to put up with high power prices and a grid that
continues to crash during summer peak loading
The people want cheaper and reliable power but are still too scared/shy to let the govt bring in nuclearDave

 

[Rive]

We just need to wait a few more years until the solar fad wears off.

I think that would be more about wind instead of solar. Solar is always in sync with the mid-day consumption peak, what makes some further room for it to grow. But wind has no such sync (not in large scale), so for its growth the base load must be sacrificed => no real nuclear option when wind is pushed.

Just for some spreadsheet exercise I've once made a calculation that ~80% non-CO2 electricity seems possible if it is nuclear+solar based. The price would be still far higher than a nuclear + fossil or purely fossil based electricity, but still not as high as any fossil + wind or, especially: mixed green + storage based scenario.

What makes a nuclear + solar option rather convenient is that the increase in price (compared to the cheapest, no PV solutions) almost entirely depends on the PV installation costs. In case PV would get really cheap, then this scenario is just unbeatable, with a lot of surplus energy available in the mid-day period.

Prerequisite is, that PV should be remotely controlled (according to demand) and: cheap.

 

[mfb]

What would be the cost of transporting it to the users? LOX is much cheaper than petrol, thus it is usually it is produced right at the use site. Needs only air and electricity.

And a plant to do so. For smaller users a truck is cheaper.

 

[russ_watters]

I think that would be more about wind instead of solar. Solar is always in sync with the mid-day consumption peak, what makes some further room for it to grow.

One issue with the way we talk about electricity is that we don't always make it clear the relationship between power and energy. For baseload nuclear running at ~90% capacity factor, the discussion is pretty much irrelevant (at current capacities). For natural gas peaking plants with low capital costs, it doesn't hurt much to build them and not run them: you save on fuel. For hydro, where you can store the energy in the reservoir, it doesn't matter either. But for solar, where the capacity factor in the USA is about 25% (and can only get worse as it expands beyond the west), it means that there are significant opportunities for solar to be wasted at even 10% of our electrical energy generation, because its capacity is 40% of our power capacity.
https://www.eia.gov/electricity/monthly/epm_table_grapher.php?t=epmt_6_07_b
https://en.wikipedia.org/wiki/Capacity_factor

The USA currently has 58 GW of installed solar capacity out of a total of about 1,100 GW generating capacity, or about 5%. From that it will get a little more than 2% of its electrical energy from solar. The peak load is about 770 GW.
https://www.seia.org/us-solar-market-insight
https://en.wikipedia.org/wiki/Electricity_sector_of_the_United_States
https://www.statista.com/statistics/187322/us-electric-peak-load-since-1990/

This is why having solar provide 10% of our electricity sounds low, but really isn't.