Energy Infrastructure - Sun Power

Introduction

Energy has been critical to mankind ever since the industrial revolution. The world's present and future depend heavily on energy so that makes it is an appealing choice for the subject of this new post.

The US consumed about 26,000 terawatt-hours of energy in 2011. This is expected to grow about 15% by 2050.

Of this total, the US consumed about 10,000 terawatt-hours worth of mostly fossil fuels to produce 4,000 terawatt-hours of electrical power with a 40% efficient process. The consumer cost of this power was about $400 billion for power generation and roughly that much again in distribution fees, taxes, etc..

Meeting this demand translates to an effective generating capacity of 457 gigawatts. As electric vehicles become more prevalent, and they will, this number will also go up by a factor of two or more.

We can get all this power for free from the sun.

We can do it with zero carbon emissions.

We can do it without any harmful wastes.

Why aren't we doing it?

My objective is to see if there are any economic or technical problems that could prevent us from doing it.

History

We started out by burning wood, then coal, then petroleum and finally moved to directly burning atomic fuel.

It took a few hundred years to discover that all of this burning has some seriously bad side effects. First we denuded the countryside, then we started dissolving our cities with the sulfuric acid produced by coal fires, then we created something called smog and finally we contaminated vast stretches of land by letting an atomic fire burn out of control in an unlikely spot called Chernobyl.

Fortunately we don't burn much wood anymore, at least not in the industrialized northern hemisphere.

There's talk of "clean coal" (pumping the carbon dioxide into underground reservoirs) but this technology is unproven on the vast scale needed to power our technological civilization.

Petroleum is the current favorite since it is relatively cheap and still plentiful despite regular dire predictions that it will get scarce. The downsides are smog, oil spills and the worldwide cost of about $3 trillion per year (nearly 90 million barrels per day in 2010: http://www.indexmundi.com/energy.aspx). We pay a big chunk of that, about 16%.

We also use 80 million acres of farmland, tens of trillions of gallons of water and trillions of pounds of nitrogen to grow corn to produce 14 billion gallons of ethanol to burn in our vehicles all the while heading toward electric vehicles. Not a great plan.

Nuclear power was once thought to be the holy grail but we could never get fusion to work. In addition, world uranium supplies are not sufficient for widespread, long term use. Finally, we also produced 70,000 metric tons of radioactive waste over 50 or so years and we have nowhere to put it.

Humans are resourceful though. We've tried water, wind, corn alcohol, tides, geothermal and solar. Water projects are on the down slope for environmental reasons. Wind is going strong but only when it blows steadily: but there are moving parts so reliability is a concern. Corn-ethanol is politically driven and not sensible if it makes us choose to either drive or eat. Tides and geothermal are extremely location dependent.

Solar has a supposedly steep implementation cost but delivers clean, free energy with no moving parts and can be easily deployed to great effect where few want to live; the desert.

Since I have freely admitted to being a fiscal conservative, this technology appeals to me since as an investment, payback comes quickly and the future returns are virtually unlimited.

President Obama was skewered for investing in Solyndra and other green technologies, and rightly so. His vision was too myopic. We need to go really big. This is something that is both worthy of a national debt and a means to pay it off.

The Plan
I will now do a paper design for a US solar power infrastructure to frame the discussion.

The design goal will be to produce four times the present consumption so as to meet all foreseeable needs. This design will also allow us to become serious energy exporters as we grow to consume it all ourselves.

So let's design for a 2.28 terawatt (four times the current 457 gigawatt plant) solar infrastructure so it will easily handle the load by the time we've finished building it.

The power grid looks like this:

US Electrical Power Infrastructure

The generating system on the left would be our solar farm. The customer loads on the right would be augmented with storage systems that allow a customer-specific portion of the incoming power to be diverted for storage to supply power at night. The transmission facilities in the middle would be upgraded to carry four times more power in a more efficient way.

Although there are a number of technologies to extract electricity from sunlight, I will use photo-voltaic for my paper design since it is most easily scaled for size, has no moving parts and is easiest to build and explain. It is also a mature technology.

First, the basics. We'll need:

• solar generation of power
• suitable land on which to put it
• mechanism to convert the direct current (DC) output of solar cells to alternating current (AC)
• improved transmission infrastructure to carry the higher power
• distributed storage mechanism to account for the fact that the sun doesn't help at night.
• manpower

Solar Generator
The sun delivers 1,000 watts per square meter at the earths equator at noon, local time.

As we move north or south away from the equator, the intensity of sunlight is diminished by absorption in the atmosphere. Similarly, as the Earth rotates, the amount of atmosphere between the sun and a solar energy collector varies.

When the weather is bad, the clouds also diminish the amount of solar energy at Earth's surface by both reflection and absorption.

For those unfamiliar with photo-voltaic systems, there are thousands of descriptive web pages.

Currently, world production of solar panels is about 60 gigawatts/year (assuming 150 Watts per square meter). At this capacity, delivering the as-designed number of square meters of solar panels will take 61 years. The current electrical power demand requirement of the US could be delivered in 15 years with this production capacity. This seems fast enough to me although the rate could improve with a little more investment in solar panel capacity.

Costs for solar cells are currently about $1/watt, excluding the inefficiencies discussed above: this is for the standard 15% efficiency. This design assumes $0.50 per watt due to volume: all of it.

Land for Generation
With all of the inefficiencies lumped together (see spreadsheet below), our initial 1000 watts per square meter becomes about 94 watts per square meter. So, for 2.28 terawatts, we need 24.3 billion square meters of solar cells.

Ideally then, we want a place with clear weather which is why the desert areas are so appealing. However, the sunniest areas are fairly mountainous and difficult to work with. The area of Texas near the western end of the Oklahoma panhandle seems ideally suited:

• more than 7,000 kW-hr/day per square meter (top map below, darker indicates sunnier)
• fairly centralized in continental US
• fairly flat terrain
• low tornado probability
• low population density (bottom map below, white space is emptiest).

Farmland in Texas averages about $1,200 per acre.

US Solar Power Availability

Tornado Alley

US Population Density

Well, 24.3 billion square meters is the same as 9,400 square miles. This requires a chunk of land roughly the size of Vermont, preferably situated in northwest Texas. Even though using arable land seems un-green, we'd no longer have to use any for growing corn for ethanol (same goes for water and fertilizer uses for this purpose). Further, the solar farm would use only 6.5 million acres as compared to the 80 million acres currently used to grow corn for ethanol: a net gain of more than 73 million acres for food growth.

The ethanol thing deserves mention for the 23 trillion gallons of water and millions of pounds of nitrogen used to grow the corn. I don't see how this can be better than solar: it still pollutes and it uses a bunch of water and nitrogen which are arguably scarce and perhaps more valuable than the ethanol.

There are valid arguments for spreading out the generation: politics, terrorists, increasing length of effective day, etc.. I'll leave those arguments to others.

Suffice it to say that the selected spot is as ideal as it gets.

DC to AC Conversion
Photo-voltaic solar cells produce direct current so we also need something called an inverter to transform the direct current to alternating current for use on our power grid. The best inverters are 98% efficient and cost about $0.35 per watt. This design assumes $0.10 per watt due to volume: again, probably all of it.

Transmission
We need to transport the electricity from the desert to where it's needed and this is also somewhat inefficient. Currently, with distributed power generation, transmission losses are about 6%.

(http://www.nema.org/Products/Documents/TDEnergyEff.pdf).

If we used copper instead of aluminum we could improve that by about 60% although oxidation of the copper is a reliability concern. If we paralleled transmission lines and/or transmitted with higher voltages (and lower currents), we could go further with more power with equivalent losses since losses decrease by the square of the voltage. Let's stick with 6% losses (94% efficiency).

The power grid needs an upgrade to handle the larger loads envisioned. The grid currently comprises some 180,000 miles of overhead cabling and some buried cabling.

US Electrical Transmission Grid

Estimates vary depending on the means used to affect the upgrades. As a rule of thumb, assume about $100,000 per mile for cabling, plus the cost of new transformers, towers, poles, testing and installation of the inverters and flywheels. Lots of work to do. About $1.5 trillion will be needed over the course of the project.

http://www.tulsaworld.com/business/article.aspx?subjectid=49&articleid=20120513_49_E1_CUTLIN461596

Speaking of transmission, I have no clue why my electric bill has more charges for non-power items than for power. The grid charges should not exceed the maintenance and insurance costs. At roughly $10 billion per year on 4,000 terawatt-hours, they should amount to no more than 0.25 cents per kWh. Below is shown a summary of my January electric bill in Connecticut. We heat with oil. This is nonsense. For 1128 kWh, we shouldn't be paying upwards of 19 cents per kWh, more than half for non-power items.

Storage for Night
Moonshine, while a delightful beverage, is fairly useless as an energy source. This means we need a huge energy storage mechanism to carry the energy collected during daytime into the night.

This will allow us to ignore the rotation of the Earth that is otherwise a large deterrent to solar power

We'll therefore need energy storage suitable for storing 0-70% of the captured solar energy for use during the nighttime hours. The wide range is needed to cover the potentially zero requirement for industry at night to a likely 70% for residential use at night.

These storage devices would be located near the customer substations; spreading them out makes sense for lots of reasons.

The two most efficient storage mechanisms that are also environmentally friendly are flywheels and compressed air. Batteries and capacitors are possibilities but these are expensive and less environmentally friendly. I've chosen flywheel technology with magnetic bearings for efficiency (85%), reliability and cost of about $1 per Watt-hour but I assume it can be reduced to $0.50 per Watt-hour for this project due to volume. An example of flywheel technology is shown below.

http://beaconpower.com/files/Beacon_Power_presentation_ESA%206_7_11_FINAL.pdf

Manpower
The current solar power workforce is over 100,000 people. To generate 20,000 times as much power will require a lot of people. We'll be spending over $30 billion/year in labor to build it: installing the solar farm and inverters, upgrading the transmission facilities and installing the storage devices. With a US average wage of $62,000, this translates to nearly 500,000 jobs for over 60 years: 30 million man-years just to build it.

By the way, the total yearly cost is about what we currently pay for the, in my opinion, useless Department of Energy ($73 billion/year).

Once it's built, a huge workforce will also be needed to maintain it.

For example, just keeping the solar panels clean requires nearly 10,000 workers just to wash it once a year to keep the efficiency up.

The transmission grid operators presently spend about $10 billion/year on maintenance and that will continue and likely increase with the proposed grid expansion.

The inverters and storage devices will also require maintenance.

Payback

The design provides a gradual ramp up of generating capacity of about 38 real gigawatts per year (60 gigawatts/year reduced by inefficiencies of the whole system). The resulting 329 billlion kilowatt-hours per year produce compounded savings on electric power generation of $32.9 billion per year (at residential prices of $00.10/kWh). The monetary break-even point happens in the 4th year (see graph below).

Power, Cost and Revenue for Solar Power per Discussion

The entire capacity of the current electrical grid would be available from solar power alone in 15 years.

At the end of 15 years, the solar plant will have generated $4 trillion in total revenue: a profit of nearly $3 trillion. Even if I recalculate to pay full price for the components, break-even just shifts to the right by about a year.

Power, Cost and Revenue for Solar Power Full Price

At the end of 30 years, the halfway mark, the farm will have already generated nearly $16 trillion in compounded revenue: nearly enough to pay off the existing national debt. At that time, it will be generating $1 trillion in annual revenue.

At the end of 61 years, it throws off $2 trillion in annual revenue and can have earned $64 trilliion over it's life: plenty of money to replace it with a new one.

Summary
With the stated efficiency of each component, we end up with:

• 94 Watts/square meter delivered to the end user (about 9.4% efficient, overall)
• 2.28 terawatts of continuously delivered power 24-7 within 61 years
• 9,400 square miles of generation - 0.25% of US landmass
• 73 million less acres of farmland used for energy production
• 40% reduction of fossil fuel usage within 15 years
• 2,000 megaton reduction of yearly carbon dioxide emissions in 15 years
• $1.2 trillion cost to replace existing electrical infrastructure in 15 years (less than DOE budget)
• $3.5 trillion more to quadruple it
• Reduction of electric bills for all Americans by 85% (based on my bill and the exit strategy)
• Increase solar production from 18GW-h to 4,000 TW-h in 15 years
• US becomes world leader in renewable energy
• US becomes a net exporter of energy
• Reduction of water use for ethanol of 23 trillion gallons per year (nearly priceless)
• Reduction of nitrogen use for corn of 4.8 trillion pounds
• 30 million man-years of new jobs over 60 years
• No more nuclear waste needed after 15 years

The design is summarized below:

In my opinion, this is something worth borrowing for.

There are no economic reasons to prevent it.

There are no technical reasons to prevent it.

There is plenty of idle manpower to do it.

The only reason I can see that prevents this is...wait for it... politics.

Those opposed would be those invested in oil, coal, gas, nuclear, ethanol and wind power; they will gradually lose 16% of their worldwide markets. The electric generation utilities would also be opposed in the short run but long-term, they will benefit by the exit strategy described below.

Exit Strategy
Once the grid is running full-blast, the government can sell enough of it to meet national demand to the utilities with the stipulation that they charge no more than an inflation-adjusted penny per kWh and that they maintain it. The federal government would also charge said utilities a penny per kWh in tax to help fund health care and social security. States would also be allowed a penny and the American people would benefit by the other 7 cents since they paid for it.

One penny per kWh equals $200 billion dollars with a 2.28 terawatt plant.

Industry would not receive preferential rates since the rates I'm discussing are already half of what industry pays now.

Grid operators would have their charges capped at an inflation-adjusted penny per delivered kWh and would pay 10 cents for every hour of outage. The same penalty will apply to the utilities. This would be incentive not to skimp on maintenance in favor of outlandish executive salaries.

A new energy export utility should also be formed to export excess energy. This could be done by converting excess energy to hydrogen and oxygen by electrolysis of water.

http://www.nrel.gov/hydrogen/pdfs/36734.pdf

The government would charge the exporters fair market value as a special tax on utilities. The process is currently only 73% energy efficient but with the extremely low cost of energy, there is incentive to improve the efficiency.

The federal government would use this income to reduce/replace taxes until the world follows our lead.

Government regulators can verify compliance by reading meters like the utilities do.