Revised and updated June 19, 2015
Review of Forbes On-line Magazine Article “Solar Energy Revolution: A Massive Opportunity”
By: Tom Tamarkin
Founder FuelRFuture & President USCL Corp
- 29.3 billion 1 square meter solar panels are required for 100% solar power in the U.S. based on current demand 24 hours a day, 365 days per year.
- 29.3 billion 1square meter panels would cover 29,333 km2 which equals 7.2 million acres, or almost all of Maryland and Delaware.
- If 1 square meter PV panels were manufactured at the rate of 1 per second, it would take 929 years to manufacture 29.3 billion panels
- The cost of a solar only approach exceeds $15.27 trillion
- To meet all energy demands for transportation, industrial, and commercial-agriculture would require 176 billion solar panels and 5,574 years to produce
- Moore’s Law is not applicable to the production or deployment of solar panels
- Increases in “solar cell efficiency” have little impact on land area to produce utility scale power
- Unsubsidized Solar has applicability in rural areas and developing countries with low population density
- Google’s Green Energy Project RE<C was canceled; “Renewable energy technologies simply won’t work; we need a fundamentally different approach; Suggest “A disruptive fusion technology…”
This paper discusses a recently published business magazine article projecting massive growth in the solar industry over the next 20 years. We have analyzed the business, scientific, and engineering backgrounds of two well-known gentlemen quoted in the article and searched for business interests that would benefit from such growth either by way of early investment and subsidy capital or long term net revenue. We have analyzed the utility industry’s need to replace an existing 440 GW of fully operational and cost effective generating capacity in light of its projected retirement of plants due to age coupled with the potential increase in demand based on partial electrification of the transportation system. We conclude with the analysis of the feasibility of powering the U.S. electricity needs by a solar-only generation infrastructure based on system components and the feasibility of extremely large volume manufacturing, capital costs and the huge land areas required.
Peter Diamandis, Co-Founder & Executive Director of Singularity University, in Moffett Field, California recently wrote an article published in Forbes on-line magazine titled “Solar Energy Revolution: A Massive Opportunity” The article starts off by stating Singularity Co-Founder and Google Director, Ray Kurzweil, “projects that the U. S. will meet 100% of its electrical energy needs from solar in 20 years.” Mr. Diamandis also states that Elon Musk, Chairman of the electric vehicle company, Tesla Motors, SolarCity and SpaceX “expects solar power to provide 50% of America’s electricity in 20 years.”
Peter Diamandis is probably best known as founder of the X Prize Foundation. In 1980 he enrolled at MIT to study biology and physics where he graduated with a degree in aeronautical and astronautical engineering. In 1989 he graduated from Harvard Medical School.
Ray Kurzweil graduated from MIT in 1970 and worked closely with the famed Marvin Minsky in the field of artificial intelligence. He is the recipient of the MIT-Lemelison award in innovation, and has received the National Medal of Technology from the White House as well as the National Inventors Hall of Fame under the U.S. Patent Office. He has received 20 honorary doctorates, and honors from three presidents and is the author of 7 books. His acknowledged area of expertise is in artificial intelligence and machine learning. In 2012 he was appointed a Director of Engineering at Google, heading up a team developing machine intelligence and natural language understanding. Google has since acquired Nest Labs which developed and sells the Nest self-learning thermostat for home use.
Elon Musk received a BS in physics from the University of Pennsylvania and a BS in economics from the Wharton School. Mr. Musk is the respected founder or catalyst of Zip2, X.com-PayPay, SpaceX, Tesla Motors, and SolarCity.
Discussion & Analysis:
All three gentlemen are well educated and extremely accomplished in their fields. Mssrs Musk and Diamandis in physics & engineering; and Mr. Kurzweil in the field of artificial intelligence, computer sciences, and IT. Mr. Kurzweil is well known as a “futurist” and has an excellent record of predicting technology development paths. All three are rock solid American citizens who have spent a life time building a better future for all of us.
These track records make people assume these predictions must be true. But are they? Should large institutional investors risk substantial capital based on these predictions? Should individual and family investors bet their retirement savings on these predictions? And perhaps most importantly, should public policy and national security be based on these predictions or is further due-diligence in order? The numbers are deceptively enticing to any business person.
Since Mr. Diamandis did not reference specific statements with relevant context, an internet search was conducted to review these “quotes” and their contexts.
We start with Ray Kurzweil and a review of business and consumer publications. Many were found quoting Mr. Kurzweil to say “in 20 Years virtually all power in America will come from Solar.” The 9billion.com news publication published an article so quoting him. The article indicates Mr. Kurzweil’s predictions are based on his “law of accelerating returns” which he derives from Moore’s Law in the semiconductor industry. Moore’s law is often used to define technology development cycles and follows:
As has been observed, over the history of computing hardware, the number of transistors in a dense integrated circuit doubles approximately every two years.
The law is named after Gordon Moore, co-founder of the Intel Corporation, who described the trend in his 1965 paper. Sometimes the time frame is shortened to 18 months based on Intel’s experience in increasing chip performance and speed primarily through the release of the next generation microcomputer chip.
such progress has led futurist Ray Kurzweil to project that solar technology will compete with fossil fuels, and will be able to provide 100% of the world’s solar energy by 2030. The basis of his projection is the continual doubling of solar power every two years for the past 20 years. IT professionals might recall the concept of “Moore’s Law” in reference to computer chips.
Mr. Kurzweil is a consummate IT professional. However he does not have a solid state physics background and has not concerned himself with the fundamental parameters surrounding Moore’s law. One is chip die size. More transistor junctions cannot be put on a chip unless the silicon die is increased within certain practicalities. This issue is limited by a fundamental matter of physics based on atomic issues having to do with molecular cross sectional diameters of atoms and the interaction with free electrons traveling across a gap, as well as interconnection issues.
In April 2005, Gordon Moore stated in an interview that the law cannot be sustained indefinitely: “It can’t continue forever. The nature of exponentials is that you push them out and eventually disaster happens”. He also noted that transistors would eventually reach the limits of miniaturization at atomic levels:
It is projected that the end of Moore’s law in terms of junction density will be reached no later than 2016. That is an altogether different issue than the ability to increase performance and speed of microcomputer chips based on design optimization and investment in process capability.
The use of Moore’s Law to describe the photovoltaic solar business in terms of market penetration, fabrication and marketing and the appropriateness of solar to replace grid level baseload power generation does not apply; in fact, it confuses and misleads people who are not skilled or studied in the fundamental science. There three principal reasons:
1. The appropriateness of solar to replace grid level baseload power generation. Solar in general, regardless of the collection system: – photovoltaics or PV, concentrated PV, concentrated solar driving conventional steam turbine generators and thermal — are extremely inefficient in comparison to their enormous size and cost. It has been noted that the earth receives more energy from the Sun in just one hour than the world’s population uses in a whole year. The total solar energy flux intercepted by the earth on any particular day is 4.2 X 1018 Watt-hours or 1.5 X 1022 Joules (or 6.26 X 1020 Joules per hour). This is equivalent to burning 360 billion tons of oil per day or 15 Billion tons per hour. However the earth is spinning sphere close to 7,925 miles in diameter at the equator. Thus a fairly small amount of energy falls on a specific surface and for only a few hours at a time. Details are provided in Solar Power Technology & Economics.
People often hear that up to 1,000 Watts of energy are available per square meter of surface area and that all of it can be converted from infrared and visible electromagnetic radiation produced by the sun
into electricity. That is a serious misunderstanding. Even Robert Muller, Ph.D. author of “ >Physics
For Future Presidents” accidentally made this mistake in his book. David MacKay uses 5-20 watts of electricity per square meter of collection surface in his landmark book “Sustainable Energy – Without the Hot Air,”.
We put an expert to the task of defining just how much electricity on average can be generated per square meter (1 meter = 39.34 inches.) The number is 37.5 watts, averaged over 365 days, 24 days a year, factoring in historical weather factors such as cloud cover, fog, etc., and in extremely well suited areas in the Southwest United States. A detailed report has been provided based on converting the current 440 GW generation capacity plus required margins with battery storage. The required amount of square land area to collect the required power is 29,333 km2 (7,248,342 acres); that is larger than the entire country of Israel and 50% larger than the state of New Jersey in the USA – or nearly equal to all of Maryland and Delaware. It also equates to a square having sides 171.3 km in length. In practicality the required area would be much larger for allowance between panels to allow construction crews access and to periodically clean the panels as dust and dirt significantly affect conversion efficiency. This requires 29,333,333,333 (29.33 billion) solar panels and 4.4 million battery modules contained in a number 40 shipping container (40 feet X 6 feet 8 feet,) covering a surface area of 130.8 km2 or a square with sides of 11.4 km with zero space between modules. This data is presented in a straightforward fashion for nonscientists in the publication “Going Solar.”
2. Manufacturing considerations. Twenty nine and 1/3 billion is a very large number of panels to manufacture. As pointed out in “Going Solar” it would take 929 years to produce this number of panels if they could be built at the rate of 1 per second. For comprehension, today’s commercially available PV panels are standardized at 1.46 square meters and weigh
about 40 pounds. Fabrication is a multistep process involving silicon crystal fabrication, cell construction, interconnection, back plane and frame. Each panel needs to be inspected, tested, and certified to meet specification.
If a manufacturing rate of 1 panel per second could be achieved, it would take 929 years to produce 29.3 billion panels one square meter in dimension. Today’s current production panels weigh approximately 40 pounds and are complicated multi-component assemblies. To be clear this analysis is based on a panel 1 square meter in size. In reality panels differ in size according to the manufacturer and customer specifications. What does not change is 29,333 billon square meters of active semiconductor solar cell collection surface area must cover a similar amount of land area exposed to the Sun.
3. Misapplication of Moore’s Law to solar cell efficiency Improvements.
The issue of solar efficiency is incomprehensible to the average person to say the least. First, available energy from the Sun’s electromagnetic radiation per a given amount of surface area is a function of many factors. This is explained in “Solar Power Technology & Economics”. Because the amount of “harvestable energy” varies drastically based on longitude, latitude, prevailing weather conditions, and day of the year a series of charts has been prepared by NREL (and others) providing a simple bottom line Watt per square meter as averaged from all these factors. This is commonly referred to as insolation.
Thus, the simple increase of solar cell efficiency does not have a proportional increase in electricity produced per square meter. In “Going Solar” the insolation number used in the analysis of a 100% solar replacement of the current U.S. generation capacity is 37.5 Watts per square meter. No amount of wishful thinking can alter this fact. Thus marginal increases in cell efficiencies have a negligible effect on the tremendous land size and number of solar panels to be manufactured. The following data sets illustrate this point.
15% “panel efficiency” This is the current state of the art for most production panels
29,333,333,333 (29.33 billion) 1 sq m panels:
29,333 km2 1 @ second = 930 Years
1,100,000,000,000 ÷ 37.5 = 29,333,333,333 sq m ÷ 1,000,000 = 29,333 km2 v29,333 = 171.3 km X 171.3 km square
“panel efficiency” This is the midpoint in published numbers for Silevo/SolarCity
20,000,000,000 (20 billion) 1 sq m panels:
20,000 km2 1 @ second = 634.2 Years
1,100,000,000,000 ÷ 55 = 20,000,000,000 sq m ÷ 1,000,000 = 20,000 km2 v20,000 =141.42 km X 141.42 km square
40% “panel efficiency” This is only achievable in complex 2 gap cells with special optics
11,000,000,000 (11 billion) 1 sq m panels:
11,000 km2 1 @ second = 348.8 Years
1,100,000,000,000 ÷ 100 = 11,000,000,000 sq m ÷ 1,000,000 = 11,000 km2 v 11,000 = 141.42 km X 141.42 km square
55% “panel efficiency”
This is the maximum theoretical efficiency based on physics.
8,000,000,000 (8 billion) 1 sq m panels:
8,000 km2 1 @ second = 253 Years
1,100,000,000,000 ÷ 137.5 = 8,000,000,000 sq m ÷ 1,000,000 = 8,000 km2 v11,000 = 89.55 km X 89.55 km square
The Shockley-Queisser limit states that the maximum solar conversion efficiency of an ideal solar cell is around 33.7% assuming a single p-n junction with a band gap of 1.34 eV.
“The Physics of Solar Cells,” Nelson, Imperial College Press, London, 2002, page 300, figure 10.9, states that the maximum theoretical efficiency of a tandem four terminal solar cell is 56%
Solar cells work by converting sun light and infrared radiation into electricity. This involves a high energy photon striking the semiconductor portion of the solar cell and transporting electrons across a band gap boundary. For a comprehensive understanding of the physics involved see “The Physics of Solar Cells”, as posted as a screen readable downloadable PDF.
For simplicity the following explanation is offered. Visible Sunlight is composed of a broad spectrum of colors which correspond to increasing photon energy levels. The lowest energy photons come from infrared merging to visible red. The highest energy electrons come from violet and ultra-violet. The following is a high resolution graph of the visible electromagnetic radiation from the Sun. (The large narrow peaks are a function of how the Sun works involving the fusion of Hydrogen and Helium with related byproducts.)
The chart below provides a specific photon energy value across the electromagnetic radiation spectrum starting with low frequency radio waves and ending with Gamma rays. The area of interest for solar cells is in the wavelength area of 800 nm to 350 nm. This represents an energy level of 1 to 1.6 electron volts. An electron volt is a very small amount of energy at 1.60 X 10-19Joules. One Joule is a Watt/second. As can be seen it takes a strong energy flux density to make the solar cell produce useful amounts of electricity!
The amount of work done per captured photon energy flux can be increased if photons of different energies could be absorbed preferentially in cells of different wavelength band gap. If the solar spectrum could be split up and channeled into photon-converters of different band gaps, then more of the solar spectrum could be harnessed. Nelson describes this in pages 298-300 in “The Physics of Solar Cells,” Impearl College, UK, World Scientific Publishing Co. Ltd., 2003-2008. Nelson’s Figure 10.6 below shows a power available from optimized one, two, and three band gap systems.
Nelson’s Figure 1.07 below illustrates one possible scheme for exploring multiple band gaps, where sunlight is split up by means of dichoric mirrors and directed on to cells of different band gap.
Nelson’s Figure 1.08 below illustrates two and four terminal configurations for tandem cells. In either case, short wavelength light is preferentially absorbed in the top cell, and longer wavelength light in the bottom cell.
A complete copy of “The Physics of Solar Cells” may be downloaded here as a PDF.
The most efficient solar cell yet produced in the laboratory is 44.7% as shown in the above graph.
Clearly Moore’s Law has no application to the use of solar cells or the production of them. The fundamental limitation is the surface area of the cell or external lenses in the case of “concentrated PV” required to intercept a specific flux density of sunlight. As shown above increases of efficiency can be made with dual cells and even three cells which have a theoretical maximum efficiency of 55%. However this requires breaking down the spectrum into discreet energy bands which are then focused on semiconductors that are spectrally “tuned” to generate maximum voltage. This requires specialized dichroic prisms or filters and lenses. It also requires exotic semiconductor materials in terms of elemental components. This technique is sometimes referred to as “concentrated PV solar.”
A more common use of the term “Concentrated PV” applies to the use of individual lenses which are used to focus or concentrate great energy flux density onto a smaller surface area of solar cell silicon semiconductor. The purported advantage of this approach is a reduction of the cost of silicon and other fundamental elements used in the semiconductor portion of the cell. This can be accomplished with lenses or parabolic reflectors. This results in a considerable price disadvantage when the cost of power per square meter is considered and the assemblies are complex and do not lend themselves to mass production; certainly not at one per second. Additional cost disadvantages of this approach are the extremely high temperature the solar cell is subjected too which must be dissipated by water cooled metal heat sinks. Whereas some advocates of this approach suggest the hot water traveling through the heat sinks has value, the fact of the matter is it does not. The water will never be hot enough to drive steam turbines for power generation and the solar sites are too far away for use in building heating systems.
Examples of a lens and a mirror concentrated PV system are shown below.
In the case of lens based Concentrated PV Panels, the use of lenses requires a separation between the lenses and the solar cells based on the focal length of the lenses. This contributes to the complexity of the structure as well as to overall weight and cost. And the fundamental bottom line is that the mirrors or lenses DO NOT increase the amount of collected sun radiation per square meter of land. If anything they significantly increase the amount of land required because of the exotic construction. As can be seen Concentrated PV is not an appropriate solution for grid level power generation.
Follow this link for an example of a government subsidized study to determine the feasibility of a concentrated Photo Voltaic solar configuration.
Concentrated Solar should not be confused with PV Concentrated Solar as it was in one popular article in the9billion site where in the last paragraph they made reference to the Gemasolar plant in Spain.
Concentrated Solar works by creating water steam pressure, or in some case vaporized salts pressure, by focusing sun rays captured by tens of thousands (or in the case of the Ivanpah project in California, 170,000+) of mirrors and focusing those beams of collected sunlight on a coil located in a tower several hundred feet high. As the liquid or salts vaporize the high pressure turns a steam generator just as in a coal or natural gas fired plant. The initial benefits were thought to be the liquid or molten salts would stay warm for some time thus “building in” natural storage capability and reducing the need for battery storage. However, experience with Ivanpah has shown this does not work and its owners recently petitioned the State of California Public Utilities Commission to allow it to produce up to 30% of its electrical energy output from natural gas. Google is a principal investor in Ivanpah as well as in a molten salts Concentrated Solar project called Crescent Dunes in Nevada. Operating experience is not yet available from Crescent Dunes.
Moore’s second law.
In the case of solar panel manufacturing Moore’s second law is of far greater significance given the huge amount of manufacturing capability needed to produce solar panels in the required multibillion quantities. In the semiconductor business which is the core of the individual solar cells on each panel, as the cost of computer power to the consumer falls, the cost for producers to fulfill Moore’s law follows an opposite trend: R&D, manufacturing, and test costs have increased steadily with each new generation of chips. Rising manufacturing costs are an important consideration for sustaining Moore’s law. This has led to the formulation of Moore’s second law, which is: “The capital cost of a semiconductor fabrication facility also increases exponentially over time.”
We have not found statements by Elon Musk providing percentage of electric power market share predictions. We have found numerous references to his vision and plans: notably, this short article of June 2014 stating that he wants to deliver 10 gigawatts watts of solar panels per year. But what does this mean? Does it mean panels will deliver 10 gigawatts of power 24 hours a day, 365 days a year to electricity users – or does it mean he wants to install 10 gigawatts of panels which are specified to deliver that amount of power under controlled STC (standard test conditions?) There is a big difference.
An examination of a Solar World Sunmodule SW 250 panel shows it to consist of 60 cells 156 mm X 156 mm producing a solar panel approximately 1.46 m2. The panel is advertised to deliver 250 Watts of electricity under laboratory STC conditions. The specification provides an IV curve– where I is current in amperes and V is voltage. Current (amps) X voltage = Watts. Their curve shows that at STC laboratory conditions when the panel is illuminated at 1,000 W/m2 it produces slightly over 250 Watts. The curve also shows that at the assumed insolation defined in Going Solar the amount of electricity is 50 Watts as defined in the insolation averages. Our precise calculation puts the true value at 54.75 Watts.
Thus the actual power generated from one panel averaged over 24 hours, 365 days, is only 21.9% of the output advertised.
Production of 10 gigawatts of power based on the STC maximum 250 Watt capability of the panels would require 40,000,000 panels to be manufactured and delivered each year for Solar City to meet its goals. At a production rate of 1 panel per second they would require 1.27 years to produce.
Production of 10 gigawatts of power based on the insolation factor of 37.5% Watts/m2 with operating panels in the real world would require 250,102,040 panels to be manufactured and delivered each year for Solar City to meet its goals. At a production rate of 1 panel per second it would require 7.93 years to produce.
This is another example of Moore’s second law at work:
SolarCity states they want to deliver 10 gigawatts of solar panels yearly. The capital required to build the process capability to manufacture at the run rate of 40,000,000 panels per year is enormous. Note this is for the solar panels only. Additional requirements apply to the electrical inverters required to convert the low voltage DC to 240-120 volts AC required by electricity users.
the number of panels required to provide 10 gigawatts of power 24 hours a day 7 days a week is much larger. The capital required to build the process capability to manufacture at the run rate of 250,102.040 panels per year is astronomical. And no consideration
has been publicly stated about the DC to AC inverters and batteries required to support 24 hour, 365 day electricity generation.
Mr. Musk originally conceived the idea of solar power to market his Tesla Motors products based on “green” PR. This white paper from the Tesla Motors web site shares that vision. The business plan was developed in 2005 on a trip to “Burning Man” in Nevada with his cousins and Mr. Musk agreed to fund what became SolarCity and remains Chairman and major stockholder.
Tesla Motor’s website indicates that its Model S has a minimum range of 302 miles based on a fully charged 85kWh battery. Representative curves of range and electricity consumption in kWhare provided on their website although there are no specific curves relating to acceleration and the slope of the road meaning grade or going uphill.
Driving up hill requires considerably more energy than static elevation driving. And it relates to the velocity or speed and the slope of the hill in combination with the amount of weight in the car. The greater the load, the higher the velocity and the greater the slope, the faster the batteries drain. Add acceleration while going uphill and the batteries are drained even faster. The relationship between mass, time, slope and velocity is a complex one requiring calculus to solve and plot.
Tesla Motors website claims that a small carport sized solar panel configuration can provide enough electricity for a typical days’ worth of driving and still contribute power to the grid. Based on Tesla’s stated range of 302 miles per charge of 85 kWh, the car would obtain a range of 36 miles assuming a home solar panel system of 11.1 square meters (119.5 square feet) of panels based on the 37.5 Watt insolation factor providing 10kWh per twenty-four hours. As can be seen it would take over 8 days to fully charge the car’s batteries assuming zero use. This is opposite Tesla’s inference that the small carport solar system can support all the car’s electricity needs apart from the power grid. The cost of a 11.1 m2 solar system with batteries rated at 10 kWh (in the event a home owner wants to be independent from the grid) would cost over $10,000.
been shown to have a direct and significant business interest in “green energy” and specifically solar. Hundreds of millions of dollars have been invested and government subsidies obtained.
Google’s Chairman, Eric Schmidt, was a campaign advisor and major donor to Barack Obama and served on Google’s government relations team. President Obama considered him for Commerce Secretary. Schmidt was an informal advisor to the Obama presidential campaign and began campaigning the week of October 19, 2008, on behalf of the candidate. He was mentioned as a possible candidate for the Chief Technology Officer position, which President Obama created in his administration. After President Obama won in 2008, Schmidt became a member of President Obama’s transition advisory board. He proposed that the easiest way to solve all of the problems of the United States at once, at least in domestic policies, is by a stimulus program that rewards renewable energy and, over time, attempts to replace fossil fuels with renewable energy. He has since become a new member of the President’s Council of Advisors on Science and Technology
Tesla has likewise been shown to have a direct and significant business interest in “green energy” and specifically solar. Hundreds of millions of dollars have been invested and government subsidies obtained.
SolarCity also has a direct and significant business interest is “green energy” and specifically solar. It too has invested hundreds of millions of dollars and obtained major government subsidies.
It is to be expected that key personnel such as Mr. Kurzweil and Mr. Musk would be passionate in their position adopted by each company.
Cost of 100% solar PV Generation
A simple worksheet showing the system component costs for the 1100 GW solar-only generation system involves over 20 years, assuming the system has a 25-year life cycle due to solar panel degradation, and requires two battery replacement cycles, based on the life expectancy of Lithium ion batteries in ruggedized application. Labor has not been calculated due to the uncertainties of installing a system in which over 900 years are required to manufacture core components. It is obvious however, that a large work force will be required to install the yearly run rate of 31.5 million panels based on a turnout of 1 panel per second.
29.3 billion 1 m2 equivalent
PV panels at $125.00 each = $3.67T
4.4 million battery modules as defined in “Going Solar” at $750,000 each = $3.3T
Steel, mounting material, and assorted electronics controls, transformers = $1.5T
Copper and Aluminum for wiring and interconnection = $50B
Land acquisition of 58,666 km2 or 14,496,368 acres at an average price of $5,000 per acre = $72,481,840,000
Steel and related construction material = $750 B
First battery change out $3.3 T = $16.59T
Second battery change out $3.3T = $20.55T (3,260 batteries per year) for 20 years
The total 20 year “overnight cost” of the system is $15.93 trillion dollars
The Utility Industry:
The United States has a current generating capacity of approximately 440 GW meaning at any point in time 440 GW of electricity are available plus a plant margin of 20% to cover maintenance, breakdowns, unplanned peak demands and other emergencies. The following chart shows estimated plant retirement based on age. The utility plant life cycle for a large 500-MW-and-up plant is 60 years meaning the plant must be designed and built to last 60 years with continuous around the clock operation and minimal downtown for maintenance.
As can be seen above the utility generation capacity is relatively stable for the next eight years. Given the regulated nature of utilities and wholesale power providers there is little incentive to invest in more plant capacity other than that required to maintain parity at 2014 levels. By 2025 a significant percentage of the nuclear generation capacity will be off-line as will much of the natural gas (and coal) generation capacity. By 2035 an additional 50 GW of plan capacity must be on-line. The utility plant licensing application to operations cycle is about 10 years. Utilities’ first choice is to have additional coal plants built to carry us through 2050 at present capacity, with a 50% increase to cover demands from the partial electrification of the transportation system. However the EPA’s opposition to coal power under the current administration creates too much risk for investors. Thus natural gas plants will be the preferred choice. The technology exists today and the cost of construction and operations fits the needed profile.
Utilities at the CEO level are not swayed by the social movement and Silicon Valley trends of “going green.” The industry is part of an annual energy industry with annual revenues in excess of three trillion dollars and is well run based on 100 years of business experience and attention to the demands of both regulators and shareholders.
In 2005, the Department of Energy in partnership with green industry firms and environmental groups issued a report partially set forth in the form of an 18 FAQ report. Roughly one half of the Q&As relate to perceived global warming based on CO2 emissions from the utility company. The chart below, produced in October 2014 indicates there has been no appreciable “global warming” over the last 18 years. It should be noted that there are serious errors in physics and the presentation of data in this Q&A. The most obvious is their statement that it will require 185,000 square kilometers to produce all electricity in the U.S. from solar and they compare that to the size of the state on South Dakota. The true number is 1/6th of that, as demonstrated in “Going Solar”.
Yet many solar companies link directly to this extremely rosy and optimistic FAQ, in so far as solar energy goes. Oftentimes, as is the case of SolarCity, this is done on an investor’s section of the web site and is meant to influence the investment community.
This detailed tutorial shows the thermodynamics of Earth’s atmosphere to have
great tolerance for carbon dioxide with no increase in atmospheric, land or ocean temperatures.
The principal drivers of the “green energy” movement today are the public and political leadership’s perception of climate change. Hundreds of millions of dollars have been invested by industry, environmental groups and foundations to mold the need for green energy in the public’s mind and hence its leadership. However over the next few years this “green fad” will likely have dissipated.
As a matter of economics, solar is immediately disqualified as a suitable material energy sourcebased on its extremely low energy flux density, as shown in this comparison of conventional fuel energy sources. (Scroll up for future fusion energy comparisons and scroll down for current green energy comparisons.) Today a 2.5 GW nuclear (fission) power plant can be built for 2.5 billion dollars. The entire US 440 GW generation capacity with 20% margin could be met with 212 nuclear plants sited on existing coal and natural gas plants without major land acquisition costs and for an overnight cost of $528 billion. The “giga solar plant” defined in “going Solar” would have an aggregate cost of over $8 trillion dollars, excluding the cost of construction, and the acquisition of roughly 59,000 square kilometers or 22,780 square miles needed to site 29.333 billion 1 m2 solar collection devices.
Replacing battery modules will cost $3.3 trillion dollars every 10 years — and operating costs as well as panel life cycle and MTBF are incalculable today. In a most likely scenario, the solar panels would all have to be replaced every 25 years due to the effects of solar radiation and weather.
The utility industry is “risk averse” in every sense of the phrase. No utility CEO in the country would support solar on such a grand scale. Today utilities embrace solar only because of regulatory demands and the positive PR value — and then, only for very small amounts of power, such as that supplied by First Solar and its Topaz project in CA, the Aqua Caliente project in AZ, and the Silver State North project in NV. These supply what the industry refers to as “peaking” levels of power and oftentimes a symbolic statement encouraged by the local politicians.
First Solar was the leader in installing large projects for grid level use. In a December 2012 RenewEconomy interview with First Solar CEO James Hughes, he made the following comments regarding utility grid level parity:
“Everyone wants to talk about “grid parity” – I’ve banned that phrase from the lexicon of First Solar. Electricity has value only at a point in time and a geographic place. There is no magic number that describes the true economic cost of electricity. You may have a tariff structure that describes it that way, but that is not the reality, and frankly, sophisticated power markets don’t operate like that. So you have to look at time of day, season and location to determine the true cost of power, and there are lots of times of day, seasons and locations where solar is economic today without subsidy. So our focus is to find those places, find those times of day, and find those market structures where we can apply ourselves.”
The government’s role, past, present, and future.
Over the last decade, the government has principally focused on the possibility of catastrophic anthropogenic global warming (AGW) or climate change caused by man’s use of fossil fuels. This focus, in conjunction with the asserted need to stimulate the economy with “green jobs,” has led to tens of billions of dollars being invested in solar research, tax credits and subsidies. Furthermore, it has shaped government policy with regard to energy policy and the EPA’s effective cap on the construction of new fossil fuel plants and a push to limit per capita energy availability of energy as expressed in the MOU signed by the EPA and the United Nations Environment Programme.
More recently, the government has taken note of other viewpoints and the connection between green energy and financial abuse as exhibited in a recent United States Senate staff report.
The EPA must be made aware of the fact that energy demand will increase significantly as companies like Tesla Motors and Nisan Leaf begin selling statistically meaningful numbers of electric vehicles. The only current solutions to meet this demand is through fossil fuel utility-scale generation plants and the installed base of operating nuclear plants. The government should be discouraging “fossil fuel disinvestment,” as it is counterproductive to the nation’s national security, industry needs, and and economic health and growth.
In “Physics for future Presidents”, Dr. Robert Muller made note of Tesla Motors and the emerging electric car. Tesla has made great strides in its battery module since that book was first released. The fact that Tesla has achieved the range it has is of merit. One gallon of gasoline is equivalent to 36 kWh of electricity. A typical 18 gallon automobile fuel tank is equal to 648 kWh of electricity; the Tesla Model S has an 85 kWh battery module. However, all those batteries must be charged from clean, abundant, affordable, and reliable electricity.
Similarly the U.S. Senate Environment and Public Works Committee has issued a report calling for Critical Thinking on Climate Change in light of new scientific findings and inaccuracies of the IPPC predictions.
Focus should now turn to the viable replacement of fossil fuels this century simply because they are finite and the national security, and financial wellbeing of the country…indeed the entire world…depends on it. This is an issue of enormous importance, yet few policy makers are aware of it and little effort is being placed on potential (non-solar) solutions.
The perception of the public has been heavily influenced by the media and by the large number of “solar companies” selling home solar systems to augment grid level power: largely subsidized by government subsidies to both the solar companies and end users through tax credits and high “feed in tariffs”.
The public is largely disinterested in science and hence lacks the knowledge to properly evaluate solar’s place, but has nonetheless embraced a false solution that has a negative cash flow solution funded by their tax dollars. That level of enthusiasm poured over when a “Solar Road” project was launched on one of the “crowd source funding sites which raised over $2 million dollars for the project’s sponsors.
The solar deception (for grid level base load electricity generation)
In Mr. Diamandis’ article he set forth his concept of the 6Ds as Digitized, Deceptive, Disruptive, Dematerialized, Demonetized, and Democratized.
Solar is indeed in a deceptive phase. Much misinformation abounds:
- In Dr. Robert Muller’s “Physics for future Presidents” well publicized book he states both directly and through a contrived “ideal student named Liz” that “there is a gigawatt of power in a square kilometer of sunlight and that’s about the same as a (small) nuclear power plant.” Non-scientists would take that statement at face value. 1 km X 1 km = 1 million square meters X 1,000 Watts = 1 billion Watts or a gigawatt. A civil engineer would say “we can build that…in time.” However, in reality you must take into consideration the fact that the Sun is only shinning 12 hours per day, plus regional insolation factors, and then add in margin for maintenance, emergencies and losses in battery charge discharge cycles. Thus all of a sudden, our 1 million square meters only provides an equivalent of grid base load power of 37.5 megawatts which is 8.5% of U.S. current on-line generation capacity. For the record, “Liz” is Dr. Muller’s daughter.
- Dr. Lewis, Dr. Tsao, and Dr. Crabtree prepared a report stating “To supply the power that the U.S. consumed in 2001 (3.24 TW) with similarly efficient solar conversion systems would require a correspondingly smaller surface area, (6) A3.24TW = A15TW · (3.24/15), = 858,792 km2 · (3.24/15) = 185,500 km2. This is roughly 1.9% of the surface area (9,631,418 km2), and 2.0% of the land area (9,161,923 km2), of the U.S. (CIA 2005).”
- Neither is correct. If Lewis, Tsao, et al, were correct it would require 185 billion 1 square meter solar panels to produce our current 440GW generation capacity 24 hours a day 7 days a week. If these panels could be produced and installed at the rate of 1 per second it would take 5,886 years to complete the fabrication and installation.
- The correct required solar collection surface area is 1,100,000,000,000 ÷ 37 sq meters, made up from 29.333 billion, 1-meter-square panels, covering an area of 29,333 km2 (7,248,000 acres, or nearly the area of Maryland and Delaware combined) or a square with sides of 171.3 km long. If these panels could be produced and installed at the rate of 1 per second, it would take 929 years to manufacture 29.3 billion panels.
It is easy to see how people can become blinded and deceived, given the media exposure of the “go solar” campaign, coupled with difficulty in obtaining scientifically correct information and then analyzing it.
It is also easy to see how corporate interests and subsidy seeking are best served by letting the ambiguities remain unsettled.
And we can forgive Mr. Musk, Mr. Kurzweil and his boss Mr. Schmidt (who took this vision all the way to the President of the United States), who in turn made available huge sums of money to fund the vision. In all likelihood they simply did not calculate the numbers associated with the manufacturing run rate, necessary land acquisition, and installation issues. It is easy to be deceived when so many zeros are involved in the mathematical calculations, and that is coupled with wishful thinking.
Both Google and Tesla are making enormous contributions
to our society and economy, which will require enormous amounts of clean, abundant, practical energy. Google is perfecting its “machine driven” car, using the research and expertise of Dr. Kurzweil in artificial intelligence and machine learning.
Undoubtedly these cars will be electric to a large extent. And Tesla Motors is increasing sales of its electric cars and its battery production.
All those cars and batteries will place an enormous strain on our electrical generation capacity.
Google is in the business of connecting people and making information broadly available to everyone. It would be prudent for both companies to look beyond solar, wind, and the other available green energy sources — and help educate the public and political establishment, promote an informed collaborative effort to define and develop the next generation of green energy, as well as exploit those currently discussed.
To Google’s enormous credit, two of its “green energy project” scientists, Dr. Ross Koninstein & Dr. David Fork authored an article titled “What It Would Really Take to Reverse Climate Change; Today’s renewable energy technologies won’t save us. So what will?” published by the IEEE Spectrum on November 18, 2014. The article concludes with a section stating:
“A disruptive fusion technology, for example, might skip the steam and produce high-energy charged particles that can be converted directly into electricity. For industrial facilities, maybe a cheaply synthesized form of methane could replace conventional natural gas. Or perhaps a technology would change the economic rules of the game by producing not just electricity but also fertilizer, fuel, or desalinated water….”
Investor’s Business daily published an on-line article titled “Google Scientists Admit Renewable Energy Can’t Work” on November 11, 2014. The article noted “…the most remarkable admission from Google is that the technology just doesn’t work — at least not now. Two of the lead scientists on the RE<C project, Ross Koningstein and David Fork, both with Stanford, wrote the following devastating critique of the future of green energy in an article posted at IEEE Spectrum: “At the start of RE<C, we had shared the attitude of many stalwart environmentalists: We felt that with steady improvements to today’s renewable energy technologies, our society could stave off catastrophic climate change. We now know that to be a false hope….”
“Google’s setbacks in green energy were even more embarrassing when the company also had to admit it couldn’t even power its own data centers with the solar paneling it had installed. According to the company statement:
“The plain truth is that the electric grid, with its mix of renewable and fossil generation, is an extremely useful and important tool for a data center operator, and with current technologies, renewable energy alone is not sufficiently reliable to power a data center.”
Try lighting up a whole city.
Not to be out done, Tesla recently announced its Powerwall product which is available in 7 kWh or 10 kWh configurations.
The U.S. Energy Information Administration states that the average annual 2013 electricity consumption for a U.S. residential utility customer was 10,908 kilowatt-hours (kWh), an average of 909 kWh per month. Louisiana had the highest annual consumption at 15,270 kWh, and Hawaii had the lowest at 6,176 kWh. However that average figure is annualized and does not allow for peak power used on days when the clothes are dried or the air conditioner is running. Furthermore a residential electricity user who averages 909 kWh per month uses 30 kWh per day which is three times the Tesla 10kWh Powerwall. And this places even more demands on sizing the solar panel array as it must generate an equivalent amount of power (30kWh) over say five to seven hours to be available to the customer over a 24 hour period.
However it gets even more complicated and expensive, depending on one’s geographic location. Why? Because many people do not live in the sunny southwest but rather live in areas of the country which may have several cloudy days in a row. On cloudy days the solar system produces a very small fraction of its normal power. Thus, one would have to have enough Powerwall batteries to provide household power for multiple days when it is cloudy and the Photovoltaic panel array would have to be sized sufficiently large to fully charge all the Powerwall batteries. Thus a $20,000 system rapidly becomes a $200,000 system. This is a very important point because no one wants to wake up with food in the freezer thawing and being unable to cook it in their microwave oven.
As explained in our “Going Solar” analysis, the solar energy industry has detailed regional information it refers to as “insolation” which correlates the actual number of radiant watts per meter squared on a yearly averaged basis. This takes into account latitude, longitude, climatic conditions and length of sunshine per day. Generally speaking the further north one goes the less radiant energy there is, due to atmospheric absorption. Likewise, areas of fog and high relative humidity preform less well.
So in reality a considerably larger battery Powerwall is required if residential electricity users are to maintain their current standard of living and “go off grid.” A publically available report on the Powerwall explains details and a review published by the AP.
And it should be stressed that the above analysis is for home power only. It does not consider charging the batteries of a Tesla or other electric car
which is described in an earlier section of this article.
On May 1, 2015, a video was published by Tesla Motors titled: “Elon Musk Debuts the Tesla Powerwall.” In the video Mr. Musk explains that Tesla company is “basically Tesla Energy” and that their mission is to replace all fossil fuels with power obtained from “the handy fusion reactor in the sky” using solar photovoltaic panels and batteries to store the power when the sun does not shine.
A Complete written transcript of Mr. Musk’s presentation is available as a PDF download
The point is made that this requires only a small land area he refers to as the “blue square.” The blue square is shown superimposed on a map and located in the Northwest portion of the Texas Panhandle. The image cannot be reconciled to the numbers we have calculated based on our analysis of the required land area.
Professor Andrew Smith of the University College London Energy Institute posted that the “Blue Square” appears to be reconciled at 10,000 km2.
Our analysis shows something very different. The fact of the matter is that 29,333 km2 of active surface area solar cells (the solid state electronic component which converts photons from sun energy into electricity) are required to provide U.S. baseload power based on 2013 generation capacity 24 hours a day, 365 days a year. This does not include the inert solar panel area nor the spacing between rows of panels to allow installation and maintenance of panels and site location for battery modules with low voltage DC to high voltage AC converters. In reality considerably more land surface area than the 29,333 km2 is required.
Near the end of the video, it is inferred that the solar panels and batteries would be installed on a distributed basis meaning on residential homes and businesses and that no large land area is needed. The article referenced and linked above points out this is cost prohibitive many times over based on today’s current electricity prices. Moreover the PV panels have a life expectancy of less than 25 years and the battery life is less than 10 years at best.
From an engineering standpoint Tesla will in all likelihood state that the best configuration is a hybrid based on the disintegration of the strong, resilient, robust, power grid regulated by public policy for the public good. Their solution would be to replace the current power grid with “micro grids” of interconnected homes using solar panels and battery packs which would average out local supply and demand. However, the power distribution system is complicated and the solar generated power must be converted from low voltage DC to much higher voltage AC power using the Tesla power pack. Such a proposal would demonstrate a lack of understanding of AC power theory and the difficulty of AC parallel circuits where power is injected at multiple points in a network. In such a configuration a tremendous amount of the home generated power injected on the “micro grid” would be lost in heat due to dynamic real time changes in power factor or reactive power considerations, and the inability to precisely match frequency and phase angles as we explain in this technical article’s “engineering challenges” associated with storage and distributed generation.
Perhaps a better use of Tesla’s “giga factory” and their management capability would be to manufacture very large battery modules needed for power grid use as discussed in the storage technologies article. That could justify investment by the utility industry and supported by the rate base under full disclosure with transparent bidding and sales contracts.
- Mssrs Diamandis, Kurzweil, and Musk are all patriotic high achieving citizens.
- All three have business reasons to promote green energy based on the perceived belief that utility CO2 power plant emissions affect the climate and that in today’s world of scientific state-of-the-art solar is the best solution.
- Tesla Motors is a great company and the electric car is a great idea. However as more and more are built, we must generate much more electricity to charge their batteries. Today, setting nuclear aside due to public perception, fossil fuel electricity is the lowest cost, and most reliable.
- Solar in fact is not an appropriate utility baseload grid level solution based on cost, required land area, operational expenses, and short life cycle.
- Solar can be effective for individual corporate and a small percentage of residential customers willing to pay the high up-front costs and long payback periods.
- Unsubsidized Solar has applicability in rural areas and developing countries with low population density and extended time before they will be connected to a power grid.
- Allowing the public to develop a false sense of security, believing solar is going to meet the country’s energy demands, is neither prudent nor responsible.
- The financial well-being of the country in two to three decades depends on energy decisions that must be made over the next 2 to 5 years.
- The National Security interests of the country are not best served by solar or even the suggestion that it may be a viable solution.
- A definite need exists for a scientific breakthrough in a very high energy flux density energy source.
- A corresponding business opportunity presents itself to an entity(ies) that are willing to take the risk and fund the science, engineering, and R&D leading to such a breakthrough and subsequent commercialization. Atomic fusion is the only realistic solution to the extent we discount nuclear fission.
- On November 9, 2006 Google recorded a presentation made by Dr. Robert Busard at their facility. Dr. Busard explained the need to solve fusion before fossil fuels run out. At 2 minutes 45 seconds, Dr. Busard makes the point that most of the very bright and talented engineers at Google do not have the physics and math background to understand fusion because Google is an IT company, and hence most of its technical staff has a computer science and IT background. The Google Talks presentation is on-line.
- Google is in the business of bringing information to people throughout the world. Google should augment its investments in solar and other current green power by bringing people, knowledge, and a crowd source like a project to “solve energy” for the world as Dr. Busard mentioned.
- Google must provide honest, accurate, non-utopian, apolitical information to the public and political establishment, so that we can make informed decisions that will best serve the vital and complex needs of our entire nation – and not merely the selfish or ill-informed interests of certain industrial, political, financial or environmentalist factions.