PV potential by US state

[mooncliff]

Re: PV potential by US state

I've posted here several times that a breakthrough in battery technology would be a huge game changer. The transitional path for transportation from hybrids->pluggable hybrids->full electric, would occur fairly rapidly with a small technology advance in batteries. It may even smooth out the Peak Cheap Oil GDP graph that iTulip anticipates in the future.

Since I live off the grid and 98% of my electrical usage is generated by photovoltaics and a small wind generator, and 75% of my space heating needs are generated from passive and active solar

Wow, that's great. If you are doing that much, that is less money flowing out of the country.

I'm sure a lot of people have had a lot of trouble with econuts. Spiking trees? What the hell is that? And there is always a lot of misguided micromanagement going on, but that should not be confused with very simple things like titanium oxide elastomeric roof, solar water heater, small photovoltaic system, which in my case would make my house zero energy, all for free. The economics of doing these things is different from what we are used to, so the calculation is not obvious. It won't work everywhere at first, but I really think the world will be very different in just 10 years.

The following plan by engineers envisions maybe 10 to 20% of energy being from solar, so it is not that solar would solve all our problems; it would just be a part of the solution. The more diversified the energy supply, the less vulnerable it will be.

A Plan to Power 100 Percent of the Planet with Renewables:
Wind, water and solar technologies can provide 100 percent of the world's energy, eliminating all fossil fuels.

http://www.scientificamerican.com/ar...energy-by-2030

[c1ue]

Re: PV potential by US state

Ray Kurzweil describes how technology goes through a period in which advance seems to be imperceptibly slow, and then goes through obvious exponential improvement. We see this pattern over and over again, and it seems to be very very robust.

Outside of semiconductors - what examples would you posit?

Several technical notes on solar:

1) Solar is an electromechanical energy converter as opposed to transistors which are a logic switching system.

The increases in transistor performance have entirely to do with the reduction in sizes (hence increase in performance) - an activity which does not affect logic operations. At least until the numbers of electrons passing with each transition is so few that quantum effects become apparent (already possible at 65nm).

Solar will not benefit from this type of activity.

Solar can only increase in efficiency: modern architecture silicon cells have only a theoretical 50% efficiency maximum with more exotic architectures havings up to 70%:

http://www.lbl.gov/Science-Articles/...olar-cell.html

Note however that increased efficiency is going to also mean increased cost.

'Old style' solar cells only have 2 layers; the more efficient cells have increasingly more layers as they are multiple junction.

From the semiconductor side, 90% of theoretical maximum is impossible to achieve outside of either very specific components or in laboratories. For actual system level it has never happened.

And indeed the maximum efficiency achieved to date for solar cells is slightly over 40%; for thin film it is less than half that:

http://en.wikipedia.org/wiki/Solar_c...d_efficiencies

This represents less than a 3x improvement over the 15% efficiency solar cells of 10 years ago.

On the cost side: as I mentioned before the solar industry - at least the silicon solar cell side - very much reaps the benefit of the semiconductor industry scale.

To put this into numbers: roughly half of the cost of a process is in the equipment. 20% is materials, and the remaining 30% is the actual manufacturing.

The material cost side won't change much as wafers are wafers. Solar can use thin film glass, but again that is a separate discussion. The actual manufacturing costs for solar are probably 1/10th for traditional silicon solar cells, but that ratio increases as more complex architectures for increased efficiencies are needed. It won't ever be 1:1 or even 1:2, but 1:4 or 1:6 is very possible - the economics is what matters.

This is one reason why the solar industry is so careful to talk about $ per watt.

Thus if the semiconductor industry represents a cost = 1 with 0.5 being the materials and manufacturing, the latter half is shrunk to 0.23 (vs 0.5) for solar. But solar doesn't use new equipment. Rather the equipment costs once fully depreciated are likely 1/4 or so, thus the 0.5 equipment cost side is only 0.125 with a net solar vs. semiconductor of 0.355 or so (35.5% relative silicon solar vs. semiconductor cost).

Of that 64.5% relatively lower cost, 37.5% is due to capital equipment cost with 27% being due to intrinsically cheaper manufacturing (less complex).

As to actual costs: The actual wafer costs are complicated by the semiconductor economics - specifically that mask costs are very high hence volume is a dominant factor - but to give an idea: a mask cost for 0.18um is around $350K. For 90nm it is $750K. For 65nm it is $1M to $1.2M.

At sufficiently high volume, the mask costs level off but the relative costs serve as a proxy for relative finished wafer cost as masks are not infinite use.

Note that solar cells are using 0.25um or likely even larger grid sizes - with correspondingly lower equivalent cost: 0.25um mask costs are around/under $100K.

Given all this, my view from this data is that while costs will go down in the short term, it is actually unclear if costs long term will significantly reduce. Any gains in incremental efficiency will be at least partially offset by losses due to capital equipment costs increasing as well as possibly increased incremental manufacturing cost due to more complex solar architectures.

Note that I mentioned previously there is a roughly 20 year maximum lifetime for this type of equipment - this affects both tool cost and manufacturing cost.

The 0.25um process is now in its 14th year.

[mooncliff]

Re: PV potential by US state

You might be interested in this little snippet

Refrigerator Needs 100 Watt-Hours of Electricity a Day

Wow! Brilliant if it works. Can't we just turn the setting to 34 degrees F? I have never had a chest freezer, but it makes sense that it is insulated much better than a refrigerator. I wouldn't care if I had to reach down to take food out. I wish they would insulated refrigerators better. Even though mine is top of the line energy efficient, I still get better performance in the winter when I crack the window next to it open just a little. That drops my electric bill about $10.

[c1ue]

Re: PV potential by US state

A Plan to Power 100 Percent of the Planet with Renewables:
Wind, water and solar technologies can provide 100 percent of the world's energy, eliminating all fossil fuels.

http://www.scientificamerican.com/ar...energy-by-2030

The little detail of that wonderful plan:

the overall construction cost for a WWS system might be on the order of $100 trillion worldwide, over 20 years, not including transmission

So a little under 10% of WORLD GDP for 20 years. And assuming no cost overruns. And ignoring transition cost issues.

Yeah, that makes sense.

[we_are_toast]

Re: PV potential by US state

You might be interested in this little snippet

Refrigerator Needs 100 Watt-Hours of Electricity a Day

Wow! What a great idea. Thanks for posting this.

When you live off the grid, you of course look for the most efficient appliances available. Refrigerators have made great gains in efficiency, but they are still electric hogs and account for about 25% of my entire electricity usage.

[mooncliff]

Re: PV potential by US state

Outside of semiconductors - what examples would you posit?

Several technical notes on solar:

1) Solar is an electromechanical energy converter as opposed to transistors which are a logic switching system.

The increases in transistor performance have entirely to do with the reduction in sizes (hence increase in performance) - an activity which does not affect logic operations. At least until the numbers of electrons passing with each transition is so few that quantum effects become apparent (already possible at 65nm).

Solar will not benefit from this type of activity.

Solar can only increase in efficiency: modern architecture silicon cells have only a theoretical 50% efficiency maximum with more exotic architectures havings up to 70%:

http://www.lbl.gov/Science-Articles/...olar-cell.html

Note however that increased efficiency is going to also mean increased cost.

'Old style' solar cells only have 2 layers; the more efficient cells have increasingly more layers as they are multiple junction.

From the semiconductor side, 90% of theoretical maximum is impossible to achieve outside of either very specific components or in laboratories. For actual system level it has never happened.

And indeed the maximum efficiency achieved to date for solar cells is slightly over 40%; for thin film it is less than half that:

http://en.wikipedia.org/wiki/Solar_c...d_efficiencies

This represents less than a 3x improvement over the 15% efficiency solar cells of 10 years ago.

On the cost side: as I mentioned before the solar industry - at least the silicon solar cell side - very much reaps the benefit of the semiconductor industry scale.

To put this into numbers: roughly half of the cost of a process is in the equipment. 20% is materials, and the remaining 30% is the actual manufacturing.

The material cost side won't change much as wafers are wafers. Solar can use thin film glass, but again that is a separate discussion. The actual manufacturing costs for solar are probably 1/10th for traditional silicon solar cells, but that ratio increases as more complex architectures for increased efficiencies are needed. It won't ever be 1:1 or even 1:2, but 1:4 or 1:6 is very possible - the economics is what matters.

This is one reason why the solar industry is so careful to talk about $ per watt.

Thus if the semiconductor industry represents a cost = 1 with 0.5 being the materials and manufacturing, the latter half is shrunk to 0.23 (vs 0.5) for solar. But solar doesn't use new equipment. Rather the equipment costs once fully depreciated are likely 1/4 or so, thus the 0.5 equipment cost side is only 0.125 with a net solar vs. semiconductor of 0.355 or so (35.5% relative silicon solar vs. semiconductor cost).

Of that 64.5% relatively lower cost, 37.5% is due to capital equipment cost with 27% being due to intrinsically cheaper manufacturing (less complex).

As to actual costs: The actual wafer costs are complicated by the semiconductor economics - specifically that mask costs are very high hence volume is a dominant factor - but to give an idea: a mask cost for 0.18um is around $350K. For 90nm it is $750K. For 65nm it is $1M to $1.2M.

At sufficiently high volume, the mask costs level off but the relative costs serve as a proxy for relative finished wafer cost as masks are not infinite use.

Note that solar cells are using 0.25um or likely even larger grid sizes - with correspondingly lower equivalent cost: 0.25um mask costs are around/under $100K.

Given all this, my view from this data is that while costs will go down in the short term, it is actually unclear if costs long term will significantly reduce. Any gains in incremental efficiency will be at least partially offset by losses due to capital equipment costs increasing as well as possibly increased incremental manufacturing cost due to more complex solar architectures.

Note that I mentioned previously there is a roughly 20 year maximum lifetime for this type of equipment - this affects both tool cost and manufacturing cost.

The 0.25um process is now in its 14th year.

OK, let's see what I can actually get next year. All those problems need to be attacked.
I think even a 1 kilowatt system will cancel my electric bill, making my total energy bill for the house zero.
Everything I have saved from the solar water heater and the elastomeric roof ($40k) makes the photovoltaics free to me... Will post something in 6 months to a year about what happens, with details.

[Kadriana]

Re: PV potential by US state

Yes, I fully agree that current hydrogen storage technology is not adequate. I was not suggesting that hydrogen storage should be used by residences. I believe the future will have a far more distributed electrical generation approach compared to the massive centralized generation we currently have. The wind farms that are popping up would be more economically viable than they already are if they had a storage capacity for windless days.

I've posted here several times that a breakthrough in battery technology would be a huge game changer. The transitional path for transportation from hybrids->pluggable hybrids->full electric, would occur fairly rapidly with a small technology advance in batteries. It may even smooth out the Peak Cheap Oil GDP graph that iTulip anticipates in the future.

Since I live off the grid and 98% of my electrical usage is generated by photovoltaics and a small wind generator, and 75% of my space heating needs are generated from passive and active solar, I am very amused by Starving Steves (no offense Steve;)), lengthy rants against Solar and wind.

My husband and I were driving in the country the other day and passed a pretty rusty substation. I'm starting to wonder how smart it is to depend on someone else for your electricity.

As a side note, Americans use way too much power. I know I'm guilty of leaving the big tv on all day for background noise. We've been watching our power usage to see how much we would need to produce if we went to alternative energy and we can easily cut back how much we use by 2/3.

[mooncliff]

Re: PV potential by US state

The little detail of that wonderful plan:



So a little under 10% of WORLD GDP for 20 years. And assuming no cost overruns. And ignoring transition cost issues.

Yeah, that makes sense.

Yes, I would rather have a free ride, but I think the free ride is over... on the other hand, I think I can get my house to zero energy for free next year... actually, I could do it now, but am waiting for details on feed-in tariff.

I went shopping one day, and I had a can of sardines in my hand, and I thought, how is it possible that someone mined the the metal, fashioned the can, sent a boat out to catch the fish, processed the fish, steam pressured the can sealed, and transported the can to the supermarket where I picked it up off the shelf... all for $1?

And of course the answer is, it is not possible without huge invisible fossil fuel burning subsidizing and externalizing the costs...

Actually, I think if we make the investments, the energy costs will start dropping around 2020, and then we will have a boom the likes of which are beyond the dreams of Avarice...

[c1ue]

Re: PV potential by US state

Everything I have saved from the solar water heater and the elastomeric roof ($40k) makes the photovoltaics free to me... Will post something in 6 months to a year about what happens, with details.

Please do.

I'd also note that the $40K you spent thus far is equal to 40 years of electrical bills for 4000 Kwh/year @ $0.25

Thus in fact what you've really done is just prepay your electrical bill - to someone other than the power company

[we_are_toast]

Re: PV potential by US state

Quote:

Originally Posted by mooncliff A Plan to Power 100 Percent of the Planet with Renewables: Wind, water and solar technologies can provide 100 percent of the world's energy, eliminating all fossil fuels. http://www.scientificamerican.com/ar...energy-by-2030

The little detail of that wonderful plan:

Quote:

the overall construction cost for a WWS system might be on the order of $100 trillion worldwide, over 20 years, not including transmission

So a little under 10% of WORLD GDP for 20 years. And assuming no cost overruns. And ignoring transition cost issues.

Yeah, that makes sense.

Let's look at the alternative. Since the capital costs of wind and solar are now only slightly higher ($/watt) than the cost of fossil fuel (or the much higher cost nuclear) mega power plants, and since all the fossil fuel mega power plants will have to be replaced eventually, the 10% of world GDP capital costs for electrical generation will be roughly the same whether it is solar/wind or fossil fuel. Looks like the only difference is that the price of sunshine and wind will remain free, while the costs of fossil fuel/nuclear will increase by orders of magnitudes.

Yep, makes a hell of a lot of sense.

[we_are_toast]

Re: PV potential by US state

Please do.

I'd also note that the $40K you spent thus far is equal to 40 years of electrical bills for 4000 Kwh/year @ $0.25

Thus in fact what you've really done is just prepay your electrical bill - to someone other than the power company

I purchased my system components from the internet and installed it myself. Total cost of roughly $15,000. Since the nearest power line is some 5 miles away, the land I purchased was at a substantial discount. Before I used my first watt I was about $20,000 ahead of the game and am on a free ride from here on out.

[mooncliff]

Re: PV potential by US state

Please do.

I'd also note that the $40K you spent thus far is equal to 40 years of electrical bills for 4000 Kwh/year @ $0.25

Thus in fact what you've really done is just prepay your electrical bill - to someone other than the power company

I think there is some misunderstanding here.

I paid in current dollars about $3,000 for the solar water heater, and $12,000 for the roof. I saved $30,000 in electricity for the water heater, and I would have had to reroof anyway because the roof was 25 years old and was getting crunchy, instead of a conventional reroof being $25,000.

So, doing the conventional thing would have been a $30,000 electric bill and a $25,000 roof, for a total of about $55,000, whereas with the solar water heater and the elastomeric roof, the actual cost in current dollars was $3,000 for the water heater and and $12,000 for the roof. So the saving is $55 minus 15, which is $40,000 SAVED, and I think I could cancel my electric bill for a $15,000 system now, without tax credits. So that is what I mean that it is for me way more than free.

[santafe2]

Re: PV potential by US state

Another piece of the energy resource puzzle.

http://greeninc.blogs.nytimes.com/20...on/#bozoanchor

If we fix the antiquated US energy grid, having distributed generation should make the overall electricity supply more robust?

I've tried to download the report a few times from here:

http://www.newrules.org/sites/newrul...files/ESRS.pdf

...but no report. I'm out of the office this week so maybe it's my connection? I've several comments but I need to review the report. Is it available at another link by now?

[fallout]

Re: PV potential by US state

This is solid analysis by c1ue.
I'm old, I was actually around for PV installations in the late 1970's and early 1980's. Costs on a per-kW basis haven't really changed much in 30 years, adjusted for inflation, despite year-after-year promises of technological breakthroughs and economies of scale. The small and incremental gains in efficiency have been offset by higher material costs, manufacturing costs, labor costs, and transportation costs.

Let's talk about reality for "average 'murkans", no pie in the sky theoretical possibilities in a laboratory setting.
I live in south Georgia, full on deep south, with 5.5 hours of "full sun" equivalent a day on average (straight off the insolation map).
Last year I had a typical run of the mill 2.1kW grid-tied system installed on the southwest facing portion of my 27 degree pitch standing seam metal roof (thus avoiding rack and mount costs). Equipment costs for a Fronius inverter, clamps, 12 panels (manufactured in British Columbia, incidentally), and associated hardware ran $10k (delivered), with professional turn-key installation from a licensed electrician another $6k. I applied and received state and federal tax breaks amounting to $6k. The system provides about 1350kw peak nominal power in use, or roughly 7-9kWhrs per day (significantly less on cloudy days, naturally). Call it 8kWhrs being generous. This represents about 1/4th of our annual usage. Local grid power from our coop is $0.10 USD per kWhr (we don't use TOD basis), so I should break even in about 34 years at the present rate. Probably won't live that long, honestly.

Note that we were the very first residential PV installation in our electric company's half-million person service area ever (bleeding edge early adopter), resulting in no end of hurdles and headaches initially (from getting a two-way meter to even getting them to continue to provide service to us, despite clear state law covering the matter).

Now don't get me wrong, I still have faith in distributed micro-power generation via photovoltaics, but at present this simply does not make sense for most people, nor can they afford it. Until that changes noticeably and measurably, PV will remain an expensive toy.

Outside of semiconductors - what examples would you posit?

Several technical notes on solar:

1) Solar is an electromechanical energy converter as opposed to transistors which are a logic switching system.

The increases in transistor performance have entirely to do with the reduction in sizes (hence increase in performance) - an activity which does not affect logic operations. At least until the numbers of electrons passing with each transition is so few that quantum effects become apparent (already possible at 65nm).

Solar will not benefit from this type of activity.

Solar can only increase in efficiency: modern architecture silicon cells have only a theoretical 50% efficiency maximum with more exotic architectures havings up to 70%:

http://www.lbl.gov/Science-Articles/...olar-cell.html

Note however that increased efficiency is going to also mean increased cost.

'Old style' solar cells only have 2 layers; the more efficient cells have increasingly more layers as they are multiple junction.

From the semiconductor side, 90% of theoretical maximum is impossible to achieve outside of either very specific components or in laboratories. For actual system level it has never happened.

And indeed the maximum efficiency achieved to date for solar cells is slightly over 40%; for thin film it is less than half that:

http://en.wikipedia.org/wiki/Solar_c...d_efficiencies

This represents less than a 3x improvement over the 15% efficiency solar cells of 10 years ago.

On the cost side: as I mentioned before the solar industry - at least the silicon solar cell side - very much reaps the benefit of the semiconductor industry scale.

To put this into numbers: roughly half of the cost of a process is in the equipment. 20% is materials, and the remaining 30% is the actual manufacturing.

The material cost side won't change much as wafers are wafers. Solar can use thin film glass, but again that is a separate discussion. The actual manufacturing costs for solar are probably 1/10th for traditional silicon solar cells, but that ratio increases as more complex architectures for increased efficiencies are needed. It won't ever be 1:1 or even 1:2, but 1:4 or 1:6 is very possible - the economics is what matters.

This is one reason why the solar industry is so careful to talk about $ per watt.

Thus if the semiconductor industry represents a cost = 1 with 0.5 being the materials and manufacturing, the latter half is shrunk to 0.23 (vs 0.5) for solar. But solar doesn't use new equipment. Rather the equipment costs once fully depreciated are likely 1/4 or so, thus the 0.5 equipment cost side is only 0.125 with a net solar vs. semiconductor of 0.355 or so (35.5% relative silicon solar vs. semiconductor cost).

Of that 64.5% relatively lower cost, 37.5% is due to capital equipment cost with 27% being due to intrinsically cheaper manufacturing (less complex).

As to actual costs: The actual wafer costs are complicated by the semiconductor economics - specifically that mask costs are very high hence volume is a dominant factor - but to give an idea: a mask cost for 0.18um is around $350K. For 90nm it is $750K. For 65nm it is $1M to $1.2M.

At sufficiently high volume, the mask costs level off but the relative costs serve as a proxy for relative finished wafer cost as masks are not infinite use.

Note that solar cells are using 0.25um or likely even larger grid sizes - with correspondingly lower equivalent cost: 0.25um mask costs are around/under $100K.

Given all this, my view from this data is that while costs will go down in the short term, it is actually unclear if costs long term will significantly reduce. Any gains in incremental efficiency will be at least partially offset by losses due to capital equipment costs increasing as well as possibly increased incremental manufacturing cost due to more complex solar architectures.

Note that I mentioned previously there is a roughly 20 year maximum lifetime for this type of equipment - this affects both tool cost and manufacturing cost.

The 0.25um process is now in its 14th year.

[Starving Steve]

Re: PV potential by US state

:rolleyes:

You might be interested in this little snippet

Refrigerator Needs 100 Watt-Hours of Electricity a Day

A reality check for pot-heads who think they can store food at 39F to 45F:

If you keep your fridge at "39F (+4C) to 45F (+7.5C)", you will end-up with food poisoning because bacteria will multiply just enough at these temperatures to create toxins. You would end-up dead. :rolleyes::eek::rolleyes::p:rolleyes::eek:

It's your choice: a few cents more per day keeping food COLD at (or under) 34F (+1C) employing energy from the electric-grid, or the certainty of food poisoning and a horrible death employing the tiny bit of home-made energy from your windmill and solar panels. :rolleyes::rolleyes::rolleyes::rolleyes::rolleyes::rolleyes: