Solar to provide 20% of energy by 2027

[J.P.M.]

So in 9 years the expectations are to triple the amount of global solar power from around 7% to 20%. Seems pretty steep clime even if prices keep going down but anything is possible.

~ 10% growth/yr. compounded annually or so depending on 9 to 11 yrs. or so. Ahh !!!, the power of compounding.

[jflorey2]

Solar is a technology that gets more efficient and cheaper with time. Solar has dropped in price much more than natural gas over the last 10 years and will continue to do so because that is how technology products progress over time.

Pretty much every person has said "this is going to get cheaper with no end in sight!" (and "this stock is going to keep going up forever!" or "this isn't a bubble - prices will just continue to rise!") has been wrong. Solar-PV is now running into some basic limitations - the price of silicon, the price of aluminum and glass for frames, copper for wiring. You can't make something cheaper than the parts it is constructed from, no matter how little you pay your workers or no matter how much you automate. We will see some small improvements, but the days of order-of-magnitude decreases in cost are over.

To assume that technological progress will stop in solar is like saying cell phones will be more expensive and not as good in five years.

And at some point cellphones will hit that limit, and not get any cheaper or better. However, due to the basic physics involved, we are a long way from that. Solar doesn't follow the same scaling factors - you cannot make a solar panel 1/4 the size and increase its power by 4 times. Again, basic physics.

[veritass]

Pretty much every person has said "this is going to get cheaper with no end in sight!" (and "this stock is going to keep going up forever!" or "this isn't a bubble - prices will just continue to rise!") has been wrong. Solar-PV is now running into some basic limitations - the price of silicon, the price of aluminum and glass for frames, copper for wiring. You can't make something cheaper than the parts it is constructed from, no matter how little you pay your workers or no matter how much you automate. We will see some small improvements, but the days of order-of-magnitude decreases in cost are over.

And at some point cellphones will hit that limit, and not get any cheaper or better. However, due to the basic physics involved, we are a long way from that. Solar doesn't follow the same scaling factors - you cannot make a solar panel 1/4 the size and increase its power by 4 times. Again, basic physics.

You wrongly compare the drop in prices in solar panels to a financial bubble. Better technology makes cheaper more efficient panels possible. Technology and manufacturing efficiencies get better with time. Obviously, the theoretical maximum efficiency is 100%, but current solar panels are at around 16 to 20%. There is theoretical minimum to cost though. There is a ton of research going on a solar and a lot of different technologies and type of panels.

Already, leading thin film solar manufacture is at 16% efficiency and their utility scale installed cost will below $1 a watt by 2017. Thin film has a long way to go. Also Solar City's cost per watt of their solar panels is around 50 cents for a 21.5% efficient module. Higher efficiency modules lower the balance of systems costs. Also,there is a lot of research is being done in perovskite, which is a lot cheaper than silcon solar panels. There is a ton of research going on out there. Obviously, the long-term trend is down.

[Sunking]

Crazy

[bcroe]

You wrongly compare the drop in prices in solar panels to a financial bubble. Better technology makes cheaper more efficient panels possible. Technology and manufacturing efficiencies get better with time. Obviously, the theoretical maximum efficiency is 100%, but current solar panels are at around 16 to 20%. There is theoretical minimum to cost though. There is a ton of research going on a solar and a lot of different technologies and type of panels.

Already, leading thin film solar manufacture is at 16% efficiency and their utility scale installed cost will below $1 a watt by 2017. Thin film has a long way to go. Also Solar City's cost per watt of their solar panels is around 50 cents for a 21.5% efficient module. Higher efficiency modules lower the balance of systems costs. Also,there is a lot of research is being done in perovskite, which is a lot cheaper than silcon solar panels. There is a ton of research going on out there. Obviously, the long-term trend is down.

Not so long ago the cost of solar panels was astronomical. But with the advances in silicon processing, they have
already dropped to be a small part of a system cost. Even if solar cells were free, system cost wouldn't drop so much.

Bruce Roe

[SunEagle]

Not so long ago the cost of solar panels was astronomical. But with the advances in silicon processing, they have
already dropped to be a small part of a system cost. Even if solar cells were free, system cost wouldn't drop so much.

Bruce Roe

Unfortunately I agree. At some point the cost of the materials will flatten out. The only savings will be in how efficiently they can be produced. More than likely that will require less humans and more automation during the assembly.

The other issue is the cost to install a solar system. While large Utility scale systems can keep the labor costs down the smaller home system will still keep $/watt up due to wages and over head for the installer.

[inetdog]

Comparisons to other countries have also indicated that a significant part of the cost of PV in the US is in permits, regulations and construction delays associated with them. Particularly for large (utility scale) projects.

[SunEagle]

Comparisons to other countries have also indicated that a significant part of the cost of PV in the US is in permits, regulations and construction delays associated with them. Particularly for large (utility scale) projects.

Yep. Most of that comes from restrictions on land use and push back from environmentalists that have issues with the intrusion into wildlife habitats.

Not that I want to see wild life disappear or worse go extinct but someone needs to make a decision between clean power or critters.

[DanKegel]

Unfortunately I agree. At some point the cost of the materials will flatten out. ...

That's true... and yet there might be technology changes that make panels radically cheaper.
Thin film is showing a lot of promise, and you never know, that might be awfully cheap to make someday.

The somewhat irrational enthusiasm for rooftop solar sure seems to be driving down the cost of solar...
and at some point, the beancounters are going to say "Damn, solar *is* cheaper", and utilities are going to swoop in and start building huge farms.
HUGE! It's going to be great!

[jflorey2]

You wrongly compare the drop in prices in solar panels to a financial bubble. Better technology makes cheaper more efficient panels possible.

Yep. And better technology makes cheaper, more efficient cars possible as well. But if we projected trends from the 1940's through the 1960's to today, we would now be driving cars with top speeds over 600mph that got 750 miles per gallon and cost $50. The reason we are not is not that research on cars has been lacking - it's that car manufacture ran into some very fundamental physical and economic limits.

Even Moore's Law, the law that says computing power will continue to double every two years, is running into some fundamental limits. Transistors the size of atoms don't work, and clock speeds faster than the speed of light divided by the distance across the IC don't work either. The research to overcome the speed-of-light issue, for example, isn't going to be as simple as the research to make a smaller MOSFET gate.

Already, leading thin film solar manufacture is at 16% efficiency and their utility scale installed cost will below $1 a watt by 2017. Thin film has a long way to go. Also Solar City's cost per watt of their solar panels is around 50 cents for a 21.5% efficient module. Higher efficiency modules lower the balance of systems costs.

Yep. But they are already a small fraction of system costs. To make big gains from here, we will have to (for example) figure out how to make aluminum, glass and copper more cheaply. And since we have been using them for centuries, we've already made most of the easy/feasible improvements in the process.

Don't get me wrong, I would love to see 10 cent per watt PV. They are simply unlikely.

[veritass]

Yep. And better technology makes cheaper, more efficient cars possible as well. But if we projected trends from the 1940's through the 1960's to today, we would now be driving cars with top speeds over 600mph that got 750 miles per gallon and cost $50. The reason we are not is not that research on cars has been lacking - it's that car manufacture ran into some very fundamental physical and economic limits.

Even Moore's Law, the law that says computing power will continue to double every two years, is running into some fundamental limits. Transistors the size of atoms don't work, and clock speeds faster than the speed of light divided by the distance across the IC don't work either. The research to overcome the speed-of-light issue, for example, isn't going to be as simple as the research to make a smaller MOSFET gate.


Yep. But they are already a small fraction of system costs. To make big gains from here, we will have to (for example) figure out how to make aluminum, glass and copper more cheaply. And since we have been using them for centuries, we've already made most of the easy/feasible improvements in the process.

Don't get me wrong, I would love to see 10 cent per watt PV. They are simply unlikely.

Here is a link to projected solar costs through 2030. Analysts disagree with you.

http://www.bloomberg.com/news/articles/2015-02-25/in-the-time-it-takes-to-read-this-story-another-solar-project-will-go-up

"The chart shows two key trends in past and projected costs of U.S. solar installations. The yellow is the cost of silicon, which continues to decline toward 30¢ to 35¢ a watt by 2020, according to Bloomberg New Energy Finance. The Agora report is even more bullish. “An end to cost reduction for power from solar photovoltaics is not in sight,” the analysts write. That holds even if solar systems see no more technological improvements, a conservative and unlikely assumption.
"

[sensij]

Here is a link to projected solar costs through 2030. Analysts disagree with you.

How does that disagree? No one said costs can't fall further, it is just that there will eventually be a limit. The article you linked has numbers that are roughly consistent with what was used in CA's net metering 2.0 proceeding, crossing $2 / W in 2027 maybe (roughly 3%-5% drop in prices annually through 2025-Attachment A & B) Lots of assumptions baked into those numbers though.

[jflorey2]

The chart shows two key trends in past and projected costs of U.S. solar installations. The yellow is the cost of silicon, which continues to decline toward 30¢ to 35¢ a watt by 2020, according to Bloomberg New Energy Finance.

That seems reasonable. Silicon accounted for 99 cents of the cost of a panel back in 2010 via one breakdown. So if you drop that to 35 cents you are at 61 cents a watt for the panel itself. (Balance is glass, EVA, junction box, frame etc) That's close to what manufacturers are achieving now, so some improvement by 2020 is likely.

But as I mentioned above, there are fundamental limits.

[Sunking]

Solar is a technology that gets more efficient and cheaper with time. .

Hogwash. Solar efficiency has not changed much in 60 years and efficiency has no real meaning, All efficiency means is the amount of area required. Prices have bottomed out and beginning to rise. Solar is a Mature Technology meaning not much more development to get out of it. Lot like a water glass or cup. Not much more you can do with it. If you had a clue what you were talking about you would already know that.

[SunEagle]

That's true... and yet there might be technology changes that make panels radically cheaper.
Thin film is showing a lot of promise, and you never know, that might be awfully cheap to make someday.

The somewhat irrational enthusiasm for rooftop solar sure seems to be driving down the cost of solar...
and at some point, the beancounters are going to say "Damn, solar *is* cheaper", and utilities are going to swoop in and start building huge farms.
HUGE! It's going to be great!

I actually hope that happens some day soon.

Dupont came close to developing a semiconductor paint that could be used to coat a material at very fast speeds. That process would create very cheap solar panels. Problem was the efficiency of that "thin film" solar pv never came around to much. But maybe the breakthrough is just around the corner.