Posts Tagged ‘Solar power’


Comparison of Arizona Nuclear and Solar Energy

December 9, 2015

Let’s compare and contrast solar energy and nuclear energy in Arizona. There is only one nuclear power plant in the state, the Palo Verde Nuclear Generating Station in Tonopah. There are several solar energy sites, so we will pick the Aqua Caliente Solar Project because it won the Renewable Energy World Solar Project of the Year category in their 2012 Excellence in Renewable Energy Awards.

Palo Verde Nuclear Generating Station

This nuclear plant consists of three reactors with with a total nameplate capacity of 3,937 MW. If these reactors ran for 24 hours day for 365 days a year they would yield 34,500 GWh (gigawatt hours) per year. The actual output is about 31,300 GWh per year (2010). This means they have a capacity factor of about 90%. Averaged over time Palo Verde yields 3,543 MW.

Palo Verde became operational in 1988 and is currently approved to operate until 2047, giving a lifetime of nearly 60 years.

Palo Verde’s construction cost was $5.9 billion in 1988 ($11.86 billion in 2015 dollars). Its operating costs for fuel and maintenance were about 1.33 cents per kWh in 2004 (1.67 cents in 2015 dollars.)

Based on an average power yield of 3,543 W and a cost of $11.86 billion (in 2015 dollars), the construction cost per watt for Palo Verde was $3.34 per Watt (in 2015 dollars).

Agua Caliente Solar Project

This 9.7 square kilometer solar energy farm has a nameplate capacity of 290 MW peak.  Its first year of full operation was 2014. If it were able to produce its nameplate capacity of 290 MW continuously for one year the energy output would be 2540 GWh. The energy output was 741 GWh in 2014, which means a capacity factor of 29%, an excellent result for solar energy. Averaged over time, this solar farm yields 84.6 MW.

Construction cost for Aqua Caliente was $1.8 billion.

Based on an average yield of 84 MW and a construction cost of $1.8 billion, the construction cost per watt for Aqua Caliente was $21.43 per Watt.


The cost per kilowatt hour of energy for either of these sources is combination of the construction cost and the operation, fuel and maintenance cost.  The longer the facilities are in operation the lower the fraction of construction cost per kilowatt hour.

The operation, fuel and maintenance cost for the Palo Verde Nuclear plant were about 1.33 cents per kWh in 2004 (1.67 cents in 2015 dollars.)  The great advantage of the Agua Caliente solar farm is that its fuel cost is zero, and we will assume for the sake of argument that its other operation and maintenance costs are also zero.

The following chart shows various costs per kilowatt hour for each of the facilities for various lifetimes.


1.  $0.0133 per kilowatt hour in 2004.  Converted to 2015 dollars.
2. 2013 energy output.
3. $5.9 million construction cost in 1988 dollars.  Converted to 2015 dollars.
4. 2014 energy output
5. $1.8 billion construction cost in 2014.
6. (GWh/year) x (number of years) x (1,000,000)
7. (Construction cost) / (kilowatt hours produced over lifetime)
8. (Construction cost per kWh) + (operating cost per kWh)

Two blocks of data are highlighted in yellow.  These are the most likely lifetime scenarios for each of the power generating plants.  The Palo Verde nuclear plant has had its license extended to 60 years.  Aqua Caliente solar farm is made from First Solar CdTe modules that have a 10 year material and workmanship warranty and a  warranty of 80% of the nominal output power rating during twenty-five (25) years.  It is reasonable to hope that it will last 40 years

There is one more thing to be considered.  We have assumed so far that the yearly output of each of these power generating stations it the same year after year.  That is not entirely correct.  Historically, the Palo Verde nuclear plant has increased its capacity factor through time as operations have become more efficient.  Whether that trend will continue is unknown.

Solar modules tend to slowly degrade with time.  The First Solar CdTe modules that are used at Aqua Caliente will likely decay at about 0.5% per year. The chart above gives a best case estimate for Agua Caliente and does not compensate for this degradation.

Based on the highlighted sections of the above chart, Aqua Caliente Solar Farm will likely cost about 2.5 times more per kilowatt hour than the Palo Verde Nuclear Plant over the course of their lifetimes.

One more point.  Aqua Caliente requires 9.7 square kilometers to generate an average of 84.6 MW.  Palo Verde Nuclear Plant generates and average of 3,543 MW.  So it would take 41 Agua Calientes to equal the power of Palo Verde.  That would require about 400 square kilometers.

Energy is the lifeblood of civilization.  The pursuit of energy abundance is the pursuit of healthier and more fulfilling lifestyle for greater numbers of people.  I present this data to help inform the choices that need to be made in that pursuit.


Units of energy: homes?

March 8, 2014

corrected 4/12/14

How many BTUs are in a kilowatt-hour?  How many barrels of oil equivalent (BOE) are in a kiloton of TNT?  There are a lot of different units of energy and power.  Which one is chosen at a particular time depends on the field and the customs of its experts.  It can get a little confusing when comparing numbers from practitioners in different fields.

It can be very eye opening to make the conversions.  For example, six sixteen watt CFL bulbs lit up for six hours will use as much energy as released by the detonation of one pound of TNT.  My preference is to convert powers to watts and  energies to watt-hours.

New unit for power

But there seems to be a new unit of power that I can’t find in any of my physics books.  Its called a “home.”  Here are some examples of its usage…

“The Tatanka Wind Farm, on the North Dakota-South Dakota border, will power 60,000 homes.”

“Limon I Wind Energy Center in Colorado is capable of generating enough electricity to power approximately 100,000 homes.”

“[E]nough clean electricity to power over 60,000 homes.”

“A 230 MW photovoltaic solar station in the Antelope Valley of California that will supply enough energy for 70,000 homes.”

“The new Copper Mountain 3 solar plant, which will be finished in 2015, will be able to generate enough power to supply around 80,000 homes.”

“Chicken Manure to power 90,000 Homes in the Netherlands!”


Ivanpah mirrors

Mirrors at Ivanpah

Brightsource’s Ivanpah Solar Electric Generating System in California is a case in point.  This is a solar thermal site that uses thousands of mirrors to concentrate sunlight to generate heat to run generators. says  the “$2.2 billion Ivanpah Solar Electric Generating System—the largest of its type in the world—will power 140,000 California homes.”  It looks like they are using a “home” as a unit of power.

What does “will power 140,000 California homes” really mean?

According to the EIA, the average home in California consumes about 7000 kilowatt-hours of electric energy each year  (most recent data, 2009).  That means 140,000 homes would use 9.8 x 108 kilowatt-hours (9.8 x 105 megawatt-hours) of electric energy per year.  I think we’re on the right track here, because the National Renewable Energy Laboratory says Ivanpah will produce 10.8 x 105 megawatt-hours per year.

But this unit of power called a “home”  is still a little misleading.  Although the average California home consumes about 7000 kilowatt-hours of electric energy per year, energy from other sources is also consumed.  The other big source is natural gas, which may be used for space heating, cooking or water heating.  If you think this is trivial compared to the amount of electricity used, think again.  The EIA document on residential energy consumption in California shows these graphs…

EIA California energy consumption

I think it is bad practice to use two mix different units for energy (kilowatt-hours and Btu) as the EIA has done with these graphs.  How many people can compare kilowatt-hours and Btu by looking a graphs?

The graph on the top left is where I got the estimate of 7000 kilowatt-hours of electrical energy per year for the average California home.  Notice that it is labled “ELECTRICITY ONLY.”  The graph on the lower left is for “ALL ENERGY average per household,” and indicates about 62 million Btu per California home per year.

How does 62 million Btu compare to 7000 kilowatt-hours?   62 million Btu translates to 18,170 kilowatt-hours!  In other words, 11,170 kilowatt-hours of energy consumed in the average California home comes from sources other than electricity.  If you find this hard to believe, look at the number of kilowatt-hours you used on a recent winter electric bill and look at the amount of energy, usually in “therms,” on a recent winter gas bill.  Convert the “therms” to kilowatt-hours and you will see what I mean.  It takes a lot more energy to heat water and air in your house than it does to light your bulbs or power your TV.  So Ivanpah really only provides enough energy to power 54,000 (≈140,000 x (7000/18,170)) California “homes.”

You might think that providing enough energy for 54,000 homes is still pretty impressive and makes a big dent in California’s energy needs.  Think again.  There are 12.5 million households in California.   So it would take about 240 (≈12,500,000/54,000) Ivanpahs to power them all.  Ivanpah covers about 16 square kilometers.  So it would take about 3600 (= 16 x 240) square kilometers to power all these households.

Building 3600 square kilometers of mirror arrays is a big undertaking, but wouldn’t it be worth it to power the entire state of California?  The problem is that it wouldn’t power the entire state of California.  Residential power consumption is only about 20% (1/5th) of California’s total energy consumption.  Far more energy goes into commercial, industrial  and transportation needs.

If we assume vast efficiencies then we might say that it only takes 2.5 times (instead of 5 times) the residential energy consumption to run the entire state of California.  With these assumed efficiencies Ivanpah would provide the total (not just residential) energy needs for the occupants of only about 22000 (≈ 54000/2.5) homes. It would take nearly 600 (≈2.5 x 240) Ivanpahs, a whopping 9000 (≈ 3600 x 2.5) square kilometers of mirror arrays, and $1.3 trillion (≈ 2.5 x 240 x $2.2 billion) to provide the average energy needs of the entire state.

Why talk in terms of “homes?”

The use of “home” as a unit of power has a warm and fuzzy feeling to it.  I guess good and caring people are concerned about “homes,” while cold and uncaring people talk about “kilowatt-hours.”  Using “homes” as a unit of power gives the impression (intentionally?) that all the energy needs of the people living in those homes are met.  It is much more impressive to say an energy project will “power 140,000 homes” than to say it will compensate for the total energy needs for the people living in 22,000 homes.

I believe this loose use of the English language and lazy, imprecise use of physical values  is used precisely because it yields more impressive numbers.


Comparing the Interstate Highway System to Scientific American’s “A Path to Sustainable Energy by 2030”

November 14, 2009

091111 November 09 SA cover 2In the November, 2009 issue of Scientific American, Mark Z. Jacobson and Mark A. Delucchi propose a plan to supply the world’s energy needs entirely by solar, wind and water sources by 2030. They conclude that the cost would be $100 trillion. My calculations show the cost to be more like $200 trillion.

This post dissects their comparison between the construction of the Interstate Highway System and their Energy system.


Interstate Highway System (2009 dollars):  $0.453 trillion
Jacobson’s and Delucchi’s Energy system (2009 dollars): $200 trillion

Jacobson and Delucchi say…

“Our plan calls for millions of wind turbines, water machines and solar installations. The numbers are large, but the scale is not an insurmountable hurdle; society has achieved massive transformations before… In 1956 the U.S. began building the Interstate Highway System, which after 35 years extended 47,000 miles, changing commerce and society.”

The Interstate Highway System is “largest public works program in history.” The concept was first approved by congress in 1944. But it was more than a decade until President Eisenhower signed the Federal Aid Highway Act of 1956. The plan evolved to building 42,500 miles of “super-highway” by 1975.  40,000 miles were completed by 1980.

The expected cost in 1958 was $41 billion. By 1995 the total construction cost amounted to $329 billion (in 1996 dollars). This translates into $58.5 billion 1957 dollars. That is not too far off from the original estimate.  Converting the $329 billion 1996 dollars to 2009 dollars gives $453 billion.

So if Jacobson’s and Delucchi’s estimate for the cost of their energy system is correct, then their energy plan would cost over 200 times as much ($100 trillion / $453 billion) as the Interstate Highway System to which they like to compare it.

If my calculations for the cost of their energy system are correct, then it would cost more than 400 times as much ($200 trillion / $453 billion) as the Interstate Highway System! And since they propose building their system in just 20 years, then it would be like building 20 interstate highway systems (which took about 30 years to build) every single year for twenty years.

Required surface area

Interstate Highway System – paved area: 3,500 km2
Jacobson’s and Delucchi’s Energy system (solar portion only): 500,000 km2

Composite interstate highway imageAnother interesting comparison is the amount of land required. The image at the left (click to enlarge) shows a spot check of interstate highway widths using Google Earth.  A liberal estimate of the average paved width of the Interstate Highway System is about 150 feet (about 45 meters, or 0.045 kilometers).  So, roughly speaking, the 47,000 mile (76,000 kilometer) Interstate Highway System paved over about 3,500 square kilometers ( 0.045 kilometers X 76,000 kilometers).

The area covered by solar panels in the Scientific American plan would be on the order of 500,000 square kilometers, or 150 times larger than the Interstate Highway System. (See calculated land required for Concentrated Solar, PV power plants, and rooftop solar, here)

Let’s rip up the Interstate Highway System and build a new one.

Jacobson and Delucchi claim that the expense of their energy system “is not money handed out by governments or consumers. It is an investment that is paid back through the sale of electricity and energy.” This is a soothing argument that overlooks an obvious fact: We already have a power energy system that pays for itself “through the sale of electricity and energy.”   

This is like pointing out that an Interstate Highway System would have great benefits for us, and then suggesting that we could reap those benefits by tearing down the system we have now and then rebuilding it.

It’s almost like swallowing poison so you can reap the benefits of good health after you recover.


Controversy over a proposal build a new electricity generation plant

July 1, 2008

The Rocky Mountain News reported on a supposed controversy over a proposal by Xcel Energy to build a new electricity generation plant powered by natural gas in Denver.  This plant would cost under $650 million dollars and have a generating capacity of 480 megawatts.  The RMN points out that critics…

“question the need for the plant, whose estimated cost today is more than $600 million, up from initial estimates of $436 million in November, due in part to escalating costs for labor, steel and equipment.

Opponents argue that renewable sources of energy, such as solar and wind, or energy conservation can substitute for an expensive new plant.”



I wonder if the opponents have checked the cost of solar energy lately.  According to the June 2008 survey results for the Solar Electricity Global Benchmark Price Indices, the cost per watt for industrial sized solar electricity installations is $4.94 per watt.  At that rate it would cost about $2.4 billion to build a 480 megawatt solar plant, four times the cost of the natural gas plant.  But, of course, the gas powered facility can operate near maximum capacity for 24 hours a day, yielding over 11,000 megawatt-hours of energy per day.  The solar powered plant can realisticly operate at its maximum power for about four or five hours a day on the average (in Colorado), and if lucky would yield maybe 3,000 megawatt-hours of energy per day.  To equal the the daily maximum energy output of the gas powered plant, the solar powered plant would actually need to have about four times the installed wattage, and would cost closer to $10 billion!  With typical silicon solar technology of today, such a solar facility would have a footprint of about 15 square kilometers, or roughly 200 times the size of Coors Field, as illustrated in figure 1, below.

15 square kilometers over downtown Denver

Figure 1.  Click on image to enlarge.  About 15 square kilometers of solar arrays would be needed to yield the same energy as a 480 kilowatt natural gas power plant averaged over a typical day.  That is approximately 200 times the size of Coors Field.  This image shows 15 square kilometers compared to downtown Denver with Coors field near the top.  Image is from Google Earth with annotation added by Tom Moriarty.


 Wind is a better bet than solar at this time, and in the long run is cheaper than gas per kilowatt-hour generated.  It would still be very expensive to install enough wind turbines to be able to match the continuous output of a gas fired plant.  480 megawatts worth of wind turbines would put out 480 megawatts of power if the wind is blowing fast enough.  But when the wind is not blowing fast enough, the the output will be lower.  A multiplicative number, called the “capacity factor” is used to calculate the amount of energy that is produced over time, versus the amount that would have been produced if the turbine had been running at its maximum output 100% of the time.   Roughly speaking, the capacity factor for wind power in Eastern Colorado is about 35%.  The capacity factor for modern gas fired electricity generation would be better than 85%.  So, in order to get the same energy as a 480 megawatt gas fired plant, you would have to install twice as much wind capacity, or about 1000 megawatts.  The realistic installation cost of wind power (with the required transmission lines, etc.) is about $3 per watt, as seen here.  So it would cost about $3 billion dollars worth of wind generation facilities to replace the $600 million natural gas powered plant.

However, even with the high construction cost, wind energy would still be cheaper per kilowatt-hour than gas in the long run.  Gas is expensive and going up, while wind is still free.  But wind has another problem.  When the wind is slow or zero, the power is low or zero.  It doesn’t make any difference how many watts of wind power have been installed when the wind isn’t blowing.  There is no power.  There must always be enough non-wind (and non-solar) power generation capacity to cover the load when the wind isn’t blowing (and the sun isn’t shining).

The folks at Xcel Energy figured this out a long time ago.  That is why our lights are not going out.  They know that wind turbines are a great asset for reducing the load on the more traditional types of power generation, but only when the wind is blowing.  That is why they are already the leading wind power provider in the United States, with over 2500 megawatts of installed wind capacity and plans for more in the future.  But they still must maintain the non-fickle conventional power sources, like gas, or the lights will start going out when the wind stops blowing.


Who can argue with conservation, if it means not being wasteful.  But be careful when some environmental activists says “conservation.”  The Rocky Mountain News article quotes the environmental activist, Leslie Glustrom expressing her reservations about moving from coal to gas.  In another recent opinion piece in the Boulder Daily Camera Glustom wrote:

“The alternative to building gas turbines to meet the summer peak is to begin using modern internet-based tools to manage the demand by cycling non-essential motors, lights, air conditioning and HVAC systems. There are a growing number of firms that develop these high-tech “demand response” systems, but, despite repeated efforts from citizen interveners, Xcel repeatedly refused to explore this powerful form of demand management.”

In other words, “citizen interveners,” (like Glustrom herself, no doubt) would like to use tools to control your use of pesky wasteful things like motors and lights.  Don’t worry, I’m sure they have your best interest in mind.


I am all for solar and wind energy.  Really.  Just click on the “ClimateSanity by Tom Moriarty” tab at the top of this page if you doubt me.  But I believe the market, rather than demands by “activists”  will ultimately lead to a better mix of renewables and non-renewables, with renewables gaining share as they become more cost effective.  As for Ms. Glustrom’s idea to “manage the demand,” I suggest a better alternative would be to have plenty of generating capacity and a varying rate scale for consumers based on the cost of generation by Xcel or the time of day.  That way, as the cost of generation varies with demand consumers can adjust their own practices and manage their own demand, without any help by Ms. Glustrom.


Get every new post delivered to your Inbox.

Join 69 other followers