Search This Blog

Thursday, January 5, 2012

Electric Vehicles – Fantasy or the Future

The prospect of vehicles powered by electricity holds a certain fascination, from the geopolitical perspective – reducing reliance on those problematic oil producing countries, the environmental perspective – no more tailpipe emissions, and the personal perspective – it would be really cool to have one.  Unfortunately, given the impracticality of a really long extension cord, most of the currently available electric vehicles (EVs) don’t go far enough on a charge to be widely practical.  So when a Silicon Valley startup announced its plans to build an all electric sports car with a range of over 200 miles it got my attention.  When the company (Tesla Motors) announced plans for a mid-priced luxury electric sedan with a range of up to 300 miles, and began taking orders, I decided to be one of the first to sign up and sent in my fully refundable $5,000 deposit back in November 2009.  It was a no-brainer.  If I didn’t like the product I could get my money back.  If it proved to be wildly popular I could probably sell my queue position for a profit.  I could even complete the purchase, be the first on my block to have the coolest new thing, and send my own personal “California howdy” to OPEC.  With delivery expected some time in 2012, I’ve had over two years to enjoy the excitement of getting a new car, an excitement that wears off quickly once the car is actually delivered.
With this new vested interest in the company I began to track Tesla and the EV industry in general.  With investment from Daimler and Toyota, a big federal loan guarantee, a successful IPO, and a sweetheart deal to purchase the recently closed GM/Toyota factory in Fremont, California, the prospect of the new Model S actually getting built went from being a possibility to pretty much a sure thing.  After a tour of the refurbished factory (at least the 10% of the factory that will initially be used to make the Model S) and a ride in a beta version of the vehicle in October 2011, my enthusiasm has not abated.  I may very well buy (or at least lease) one of these things.
Fortunately, there is still time for my normal skepticism and rational pragmatism to decide if there really is a future for the electric car industry and whether this automotive startup will get thrashed by the big boys, leaving me with the latest version of the failed automotive startup.  So it’s time to take off the rose-colored glasses and take a hard look at this industry. The primary concerns voiced by EV skeptics are range anxiety, potential electric grid impact and a minimal impact on Greenhouse Gas emissions.  Let's look at all three.

Range Anxiety

Most EV skeptics point to "range anxiety" as trumping potential consumer interest in switching from internal combustion engines (ICEs).  They point to consumer surveys listing it as a significant barrier to claim that pure EVs will never catch on and to justify hybrids or "assisted EVs" such as the Chevy Volt or Fisker Karma that have on-board ICEs to recharge batteries.  With pure EVs like the Nissan Leaf having maximum range under 100 miles, the skeptics have a point.  Will the Telsa’s 300 mile range be enough to overcome the anxiety or will something else be needed?

Companies like A Better Place would deal with range anxiety by building automated battery swapping stations that turn stopping for charge into a five-minute refueling process.  Besides overcoming the range anxiety concern, this business model could provide a new paradigm for battery ownership, and even provide an additional opportunity for profit if the charging stations can be configured to provide energy storage capability to support the electric grid.  But even though the Model S will have a “quick swap” battery, Tesla is not planning on participating in the Better Place program.
Instead, Tesla is planning to install a series of 90 kW fast charging stations in key locations such as along I-5 between San Francisco and Los Angeles .  These would provide a 150 mile charge in 30 minutes or so, a more reasonable turn-around than the six to eight hours typical required to recharge at standard charging stations.  More importantly, however, the 160 to 300 mile range of the Tesla Model S means that for a very large percentage of trips there will be no need to stop anywhere to refuel, just plug in when you get home.

Electric Grid Impact

Widespread adoption of EVs would reduce oil consumption and increase electricity consumption, begging the question of whether the impact of the increased electric consumption would create problems for the grid.  A little analysis should provide some idea of where the tipping point may be.  Consider California:
1.     There are roughly 32 million motor vehicles registered in California.
2.     On an annual basis Californians use about 242,000 GWh of electricity and about 24.5 billion gallons of gasoline.
3.     At 20 mpg, these vehicles travel an average of 15,000 miles per year, let’s say 50 miles per day.
4.     EVs have a range of ~3.5 miles/kWh.  Thus each one would require about 15 kWh/day or about 4,500 kWh/year.
5.     Thus, if 1% of California’s vehicle fleet was replaced with EVs, and traveled the average number of miles, total electric consumption would increase by 1,440 GWh/year, or less than 0.6% and gasoline consumption would decrease by 245 million gallons/year.
6.     Most of the EV charging would take place at night when electric rates and demand are lower.  In the extremely unlikely event that all of the EVs were recharged using high capacity (20 kW) chargers at the same time during peak usage periods (noon to 6 weekdays), they could increase peak demand by 6,400 MW, about 10% of the current peak.

Based on these numbers, even a 10% adoption rate for EVs (3.2 million) would have a fairly minor impact on overall electric consumption, provided that most recharging is done at night.  However, there could be localized problems with neighbors keeping up with the Jones’s.  Most residential service transformers have a maximum capacity of about 25 kW and serve several residences.  If more than one chose to recharge at high capacity (20 kW each) at the same time, they could overload the transformer, which is not a good thing.  Some kind of coordination would probably be needed to make sure this does not happen.  Fortunately, through a combination of special rates for EV charging and development of smart grid technology, coordination should be a viable option.

Greenhouse Gas Emissions

Another area of skepticism involves the claim that widespread EV adoption would reduce greenhouse gas (GHG) emissions.  The question is whether the tailpipe emissions avoided by EVs are greater than the increase in smokestack emissions from power plants producing that extra electricity.  The simple answer is – pretty much.  Once again, let’s look at the numbers:
·      Combustion of one gallon of gasoline produces about 20 pounds of CO2
·      Natural gas combusted to make electricity produces between 0.8 and 1.2 pounds of CO2 per kWh generated (depending on efficiency of power plant)
·      Burning coal to make electricity produces about 2.1 pounds of CO2 per kWh.
·      Other sources of electricity create virtually no CO2 emissions.
·      According to the Energy Information Administration, 45% of electricity in the US is generated from coal, 24% from natural gas and the rest from non-GHG sources.  Generation of 1 kWh thus produces 1.185# CO2 (Assuming 1#/kWh for natural gas)
·      According to the California Energy Commission, 8% of electricity consumed in California comes from coal, 42% from natural gas, 12% is unspecified and the remainder is from non-GHG sources.  Depending on the makeup of the unspecified portion, generation of 1 kWh produces between 0.59 and 0.84 pounds of CO2.  Let’s use 0.75. 
Putting it al together:
·      Driving 100 miles in an ICE vehicle produces 100# of CO2 at 20 mpg or 50# at 40 MPG.
·      Driving 100 miles in an EV requires 29 kWh which produces 34# of CO2 based on the national average and 22# in California. 
The bottom line is that EVs do produce less CO2 than the gasoline-fueled vehicles they replace.  Any action that reduced GHG emissions from power plants (like replacing coal plants with renewable resources or nuclear power) would further reduce GHG from EVs.

Are EVs Ready for Prime Time?

One mistake EV skeptics make is to consider the EV to be conventional vehicle with an electric motor – much like early cars were characterized as horseless carriages.  In fact, EVs are not subject to the same limitations as ICEVs and driving one is a much different - and superior - experience.  From a design perspective, losing a large, hot, noisy engine and drive train provides huge amounts of flexibility.  For example, while retaining the look of a sedan, the Tesla Model S will have 60 cubic feet of storage space, comparable to a mid-sized SUV.  Without the constraint of an engine compartment, who knows what basic design changes will be made.  Driving will also be a different experience.  An EV like the Tesla accelerates faster and more smoothly than an ICEV with a 0-60 mph time of under 6 seconds.  And thanks to the electric motor's flat torque curve, the same acceleration rate is available at any speed through the one-speed transmission.  Use of regenerative braking will mean a lot less need to switch from the gas pedal to the brake, reducing brake wear and maybe the prospect of accidently driving through shop windows (officer, I swear I was pressing the brake).  Electric motors have vastly fewer moving parts (1) than an ICE (~200), which means less that can go wrong.  Lack of combustion means fewer maintenance parts (filters, fluids, oil, etc) to replace and a lot less maintenance to perform.  No exhaust pipe means no smelly pollution, no catalytic converter, and don't forget - no more gasoline, ever.  EVs are so quiet that they may need to be equipped with some kind of sound-making device so blind or distracted pedestrians don't walk in front of them.  (Picture being able to download your favorite exhaust sound app for your car.)  In addition to the performance advantages there is the fact that a fill up from empty (which you can do in your garage) will add about $10 to your electric bill versus $30-$50 and a trip to the gas station.  It will also be possible to pre-warm or pre-cool an EV remotely so it won’t necessary to get into a freezing or sweltering vehicle that has been parked out in the weather.  Over time as EV design evolves, they may look and feel less and less like the old fuel burners they replace.  It may take a while, but EVs will ultimately replace ICEVs just as cars replaced horse-driven carriages.  And when it’s done no one will regret the reduction in tailpipe emissions, just like no one misses streets covered with horse exhaust.

As far as whether Tesla itself will be hugely successful, that’s harder to tell.  So far, everything appears to be going according to plan.  The Model S was promised in early 2012, and initial deliveries are expected to begin in the summer.  While options can easily get the price close to six figures, the base model, with a 160 mile range, is still priced at the $57,400 price originally promised.   As critics point out, the automobile manufacturing business is highly competitive and includes a number of well-entrenched vertically integrated international players.  Certainly their R&D departments can come up with superior designs, their manufacturing expertise can build more cheaply, and their dealer networks can sell Tesla into the ground.  From the perspective of the automobile industry, Tesla doesn’t stand a chance.  On the other hand, car companies base their EV designs on their existing ICE models.  Being in the position of designing an EV from the bottom up allows Tesla to optimize the design rather than adapting to existing vehicle limitations.  This is shown in their use of a flat battery pack for the Model S located under the passenger cabin providing stability, structural integrity and easy access.  Their approach to sales and design is more like Apple than GM, and considering that consumers are more comfortable buying i-Pads than Chevys that could be a plus.  And then there’s the fact that Tesla stock is being heavily shorted by skeptics, a sure sign of value for us contrarians.  Bottom line – I wouldn’t count them out.  (I have no holdings in Tesla.)

Friday, August 5, 2011

Hey Grid, What Makes You Think You're So Smart?

On July 1, 2011, Pacific Gas and Electric (PG&E), Southern California Edison (SCE) and San Diego Gas and Electric (SDG&E) submitted their Smart Grid Deployment Plans as required by the California Public Utilities Commission.  At 290, 178 and 391 pages respectively, the plans are definitely weighty.  They also contain lots of attractive graphics, informative statistics, and interesting projections.  They estimate an implementation cost in excess of $6 billion (on top of the more than $1 billion each for PG&E’s and SCE’s Smart Meter deployment) to achieve a set of vaguely defined benefits that they estimate to be only slightly greater than the cost.  The smart grid benefits include:
·      Safe and reliable integration of renewable energy resources.
·      Integrating electric vehicle charging into grid operations.
·      Enhanced demand response.
·      Customer empowerment.
·      Improved grid reliability – reduced outages.
·      Automated and improved grid management.
·      Foundational and cross-cutting utility systems, facilities and programs necessary to continuously improve application of new smart grid technologies.  (Whatever that means.)

Behind the plans is the implication that there are massive future benefits that have not yet been identified – Smart Grid killer apps if you will – that will do for the electricity industry what advances in cell phone technology have done for the telecommunications industry.  That promise of future market transformation and a wonderful new world brought about by the new and improved Smart Grid is the underlying driver behind this process.  Otherwise, it is just boring utility infrastructure stuff.  The problem is that the platform for developing this exciting future is being crafted by utilities and their regulators, a combination not known for innovation.  Google and Microsoft, entities with a somewhat better reputation for high-tech innovation have recently terminated their nascent Smart Grid programs.  What’s the deal?

The problem is simply that the utility industry does not and cannot undertake the kind of risk-taking and innovating needed to make the Smart Grid anything more than an incremental improvement in utility operations.  It has nothing to do with the quality, intelligence, and forward thinking of utility management and everything to do with the regulatory compact under which utilities and their regulators operate.  In other words, it’s all Samuel Insull’s[1] fault.  It was Insull who promulgated the idea of natural monopolies to avoid the costly waste of duplicate electric systems and the use of regulated rates to protect consumers.  Under this simulated market model, utility companies have their regulators determine whether investments proposed and expenses incurred are just and reasonable and issue Certificates of Public Convenience and Necessity to approve proposed investments.  Utilities are then allowed to recover their costs and a return on equity for their capital investments.  Thus, utility focus is not on offering consumers the most value or developing competitive advantage through innovative products, but on building ratebase and demonstrating to their regulators that their investments and expenses are reasonable and thus worthy of recovery through rates.  In other words, utility earnings are a function of how much stuff they can convince regulators it is reasonable for them to own.  This is not a recipe for the innovation and risk-taking required to develop new and better products or services or for finding new needs or wants that The Smart Grid can fulfill.  It is, however, a great way to build ratebase in new ways and to pass the cost to consumers based on potentially illusory – but reasonable – anticipated benefits.  Up side may be limited to an authorized rate of return on equity, but even failed investments or massive cost overruns can be charged to consumers provided they are found to be just and reasonable.  Utilities have no incentive to think outside the box, but plenty incentive to make sure the box is very well built using the finest, most reasonable, materials.  Is it any wonder that Google and Microsoft decided to take a pass?

What can we expect from utility Smart Grid activities?  The most likely thing is capital investments that reduce operating expenses.  This is consistent with the initial Smart Grid investment – the Advanced Metering Infrastructure, aka smart meters.  While utilities wax poetic about empowering customers and facilitating demand response, the primary benefits of the new metering infrastructure are eliminating the need for meter readers and the getting the ability to turn electric service on and off remotely.  Both replace pass through expense items (employees) with rate based investments in meters.  On top of that, California utilities will be able to continue to get a rate base return on the old – not smart – meters that are being replaced.  They were, after all, a reasonable investment when purchased.  Other Smart Grid investments in distribution and transmission automation will be justified based on things like reduced outage rates and durations - soft benefits, but reasonable. 

The bottom line is that as long as The Smart Grid is in the realm of utility companies and regulatory agencies its impacts will be limited and the transformative “killer apps” will remain nowhere to be found.  Remember, when cell phone licenses were issued in the early 1980s, at least two licensees were allowed in each area and only one could be a regulated telecommunications company.  It was this competition, not the reasonableness of cell phone services that produced the innovation and amazing success of the industry.  We have a long way to go before that kind of competition comes to the electric utility industry.  But at least now I can find out how much electricity my house uses each hour, information I never realized I needed before, but may soon be able to access from my Smart phone.

[1] Insull started as Thomas Edison’s private secretary in 1881, developed the regulated utility industry model into a thirty state empire, was indicted and tried for fraud in the 1930s only to be acquitted by a jury that needed only two hours of deliberation to reach their verdict. 

Friday, July 15, 2011

Googling Electricity

Google recently announced that it would be providing a $280 million equity investment in SolarCity to support SolarCity’s residential solar lease program.  SolarCity claims that the cost its 20 year solar PV system lease would be lower than the electricity bill savings over the life of the lease.  The prospect of a rooftop solar system providing electricity at a lower cost than the local electric company is a potential game changer that I had to investigate.  I was surprised to find out that by putting a leased solar system on my rooftop I can get electricity for 20 years at a fixed price of about 12¢ per kWh.  How could this be?  Does it portend the beginning of the end for the electric utility industry?  Was Jerry Brown aware of this when he made a campaign promise of 12,000 MW of distributed generation in California?  So I decided to sharpen up the spreadsheet and find out.

How Does It Work?
There are currently at least three companies offering residential solar leases or power purchase agreements in my area.  They all offer variations on the same concept:  a zero down payment ten or 20 year lease with payments (per month or per kWh) that increase by a fixed percentage (2.5 to 4%) each year.  The lease can be converted to a fixed payment for the entire term by making an up front payment of 5% or more of the net purchase price, or could be entirely prepaid.   The effective annual interest rate for deferring payment is over 13%, which takes a serious bite out of the savings and makes prepayment attractive.  The cost of the prepaid 20 year lease came out to under $3.50/watt DC, including a 20 year equipment warranty and a meaningful[1] annual production guarantee.  The maximum cost of electricity (payment / total guaranteed production) is 13.2¢/kWh, compared to my current average electricity rate of 17.5¢/kWh, a price which has increased an average of 5.3% annually over the last 10 years.  The utility’s net metering program, which credits peak period (daytime) production at higher rates, results in an actual bill reduction of 27.5¢/kWh, making the solar investment look even better.  Should utility rates continue to increase at 5.3%, the savings would average almost 48¢/kWh.

Is It Really That Cheap?
This can’t be right, there must be hidden subsidies involved somewhere.  In fact, the difference between the total cost of the system and the prepaid lease cost is about $2.70/watt – over 40% of the total system cost of $6.20/watt.  And this is without attaching any value to the performance guarantee or the 20 year warranty, which includes anticipated replacement of the inverter after a dozen years or so.  The reduction includes the California Solar Initiative (CSI) - $.30/watt,  a 30% Federal investment tax credit - $1.76/watt, and lease savings - $.64/watt.  How can it possibly cost less to lease a system for 20 years (after which they either give it to you or uninstall it and return your roof to its pre-solar condition) than to buy it?  Two things – depreciation and RECs.  Google – the equity investor – gets to depreciate its entire cost of the systems it leases in the first year, which translates to a 30% income tax reduction off the top.  Assuming a 100% equity investment and all the other subsidies, Google’s net cost of the system is $2.16/watt.  That leaves $1.34/watt for operating cost and profit.  The value of the Renewable Energy Credits (RECs) is a bit more difficult to estimate.  Values from 0.5¢/kWh up to the California PUC’s price cap of 5.0¢/kWh have been suggested, which over 20 years equates to between $.13 and $1.30/watt.  And, of course, it will be much easier for Google to extract that value from $280 million in solar investment than it would be for anyone with a small residential sized system.  So, Google gets a healthy return on its green investment, I get to save money on electricity while demonstrating my green chops, and SolarCity gets to create jobs and support US-made solar equipment.  Everybody wins.

So Who Pays?
Obviously, if I’m paying $3.50/watt for a solar system that costs $6.20/watt to install and SolarCity and Google are profiting, someone else is picking up the tab.  For that we can thank the American taxpayers and California utility customers.  Federal tax credits and accelerated depreciation cover more than half of the retail cost of the system.  In a less “climatically correct” environment these would be characterized as tax loopholes that increase the federal deficit.  The CSI rebate, one component of a 1.5¢/kWh “Public Purpose Program” charge on California electric customers, currently covers 30¢/watt (it was once $2.50/watt).  Net metering – the ability to be credited for daytime solar generation at high peak period prices (26.5 to 48.5¢/kWh) while being charged for night time usage at off peak rates (9.3 to 31.6¢/kWh) – means that every kilowatt-hour my leased solar system generates at a cost of 13¢/kWh reduces utility charges by 27.5¢/kWh, while saving only 6-10¢/kWh in fuel costs.

Is It Worth It?
Advocates of renewable electricity cite energy independence, climate change mitigation and job creation as reasons for investing in (or subsidizing depending on your perspective) technologies like wind and solar.  As I pointed out last year here, the United States uses very little oil to make electricity.  In California the marginal fuel for power production is natural gas, little of which is imported.  As a source of GHG reduction, solar costs about $187/Tonne of CO2 equivalent.  The renewable industry does appear to create jobs, however, and does so without competing with other non-government industries.  I guess we should just think of it as a 21st century version of public works projects.  Yes, I did indeed decide to buy 20 years of electricity up-front.  Just doing my part for the economy.  

[1] You are paid 18.9¢/kWh escalating at 3.9% annually for production under the guaranteed level.

Tuesday, June 7, 2011

Know More Nukes

It seems pretty likely that the Fukushima nuclear plant accident is going to send ripples (if not tsunamis) through the nuclear power industry.  Unfortunately, as is so often the case, the ripples are likely to be misguided, focus on the wrong issues and recommend untenable solutions.  Such is the controversy over nuclear power.  Rather than focus on the emotionally-laden issues that tend to drive the nuclear debate, I’d like to look at some of the practical issues and tradeoffs that characterize nuclear power.

Nuclear energy was developed after World War II as a peaceful use for the devastating power of atomic fission.  It was based on the simple idea of using the heat generated by controlled fission reaction to make steam that could be used to drive a turbine and make electricity.  Because it does not rely on the combustion of any fuel, nuclear fission was an ideal fuel for ships – particularly submarines, which can remain submerged almost indefinitely.  When used to operate power plants, nuclear energy was characterized in the 1950 and 1960s as providing electricity too cheap to meter.  It has substantial advantages over other electricity sources, particularly coal.  Characterizing coal-fueled generation as “burning dirt” is not far from the truth.  The table below compares the annual requirements for a 1,000 MW nuclear generator versus the same size coal plant (burning Power River Basin coal), both running base load, about 90% of the time:

Fuel required
5,000,000 tonnes
27 tonnes
GHG output – CO2 equiv
7,884,000 Tonnes
NOx emissions
14,000 Tonnes
Radiation released
490 person-rem/year
4.8 person-rem/year

That’s right, a coal power plant requires over 600 tons of coal per hour while a nuke needs to be refueled once every 18-24 months.  It’s no wonder that nuclear power plants might have been seen as an incredible boon to the industry.  But, of course, none of the nuclear opponents are suggesting replacing the nukes with coal plants.  Instead, the solution is usually to build more renewables, they don’t pollute, the fuel is “free,” and they create more jobs.  The renewable resources of choice are primarily wind and solar.  Time for another reality check.

Wind Power

Wind turbines are probably the most widely recognized renewable resources.  They must be located in areas where sufficient wind is present and only operate when the wind is blowing.  Even sufficiently windy areas can only support generation about 30% of the time.  As a result, to replace a baseload nuclear generator of 1,000 MW, would require over 3,000 MW of wind generation plus 1,000 MW of storage that can operate the 70% of the time when the wind is not available.  Wind turbines require about 30-50 acres/MW, so 3,000 MW would require about 120,000 acres or 187.5 square miles.  This is in addition to the storage facility, which with currently available technology, would need to be a pumped hydro storage facility which would also have a substantial footprint and cost.

Solar, Maybe?

Solar power is another favored alternative to nuclear power.  It has the same kind of limitations as wind, though it is possible to have storage incorporated into a solar resource (by using concentrating solar thermal and molten salt storage rather than photovoltaics).  But here again, many more MW would be needed to produce the same energy as a baseloaded nuke.  Solar can only produce energy when the sun is up and high enough in the sky to be collected.  A 25% capacity factor is typical for middle latitude installations, so 4,000 MW of solar would be needed to provide the same energy as our 1,000 MW nuclear plant.  Solar plants need about 7 acres per MW, so about 44 square miles of land would have to be covered by collectors to be displace a single nuclear unit.  Because of the negative impact of cloud cover, these plants would be best sited in relatively sunny areas like the desert. 

Natural Gas Generation

The most likely replacement for nuclear power is probably natural gas combined cycle generation.  It is a fossil fuel that is combusted in the generation process, but because combined cycle generation is more efficient than steam cycle generation (<7MMBtu/MWh versus 10MMBtu/MWh for coal) and natural gas is less carbon intensive than coal (117#CO2/MMBtu versus 213 for coal), it produces about 40% as much CO2 per unit of electricity as coal.  Natural gas is much cleaner burning than coal and delivered via pipeline rather than unit train, gas plants are more scalable than coal or nuclear generation.  Thanks to advances in drilling technology that are allowing access to shale gas (yet another anathema for environmentalists), natural gas availability is increasing and prices are remaining fairly stable.  The United States could replace its entire nuclear fleet with gas-fired generation and increase total gas consumption by less than 25%, a significant increase but a potentially viable one.  Besides the fact that this conversion would increase GHG emissions by about 300 million tonnes per year, if the gas were used instead to replace coal power plants, it would reduce GHG emissions by about 450 million tonnes per year.

Distributed Generation

For those who prefer to think outside the central power plant box, distributed generation is the answer.  A combination of rooftop solar with storage for residential loads could have some promise, though it would be a major undertaking.  A fairly large residential solar system would cover about 800 square feet of roof and produce a maximum of 10 kW.  Being fixed panels, these systems would produce less energy than a centralized tracking system, with approximately a 20% capacity factor.  That would mean 5,000 MW to replace a baseloaded nuclear plant.  That would be about 500,000 rooftops.  Local battery storage could be accomplished with a battery pack about the size of one used for an all electric car, about 70 kW.  To replace the entire 101,000 MW of nuclear generation in the US would require solar panels on 50 million roofs with a comparable number of electric car batteries.  Should electric car batteries and photovoltaic panels continue to drop in price, this could become a viable option – in a decade or two.  By piggybacking on electric vehicle development, this approach could actually significantly reduce reliance on imported oil.  Should fuel cells ever become a cost-effective alternative, they could also prove to be a game-changer.

The Real Problems with Nuclear Power

When they work the way they’re supposed to, nuclear power plants are very impressive.  They don’t pollute the air, don’t create greenhouse gases, require virtually no fuel, can fit in a fairly small space, and like to run flat out all the time.  They put less radiation into the atmosphere than coal and produce vastly smaller quantities of waste.  Their primary problem, when operating as designed, is the large amount of heat that must be removed from the process. Nuclear plants produce steam at a lower temperature and pressure than generators that rely on combustion can produce.  As a result, more lower temperature heat must be removed from the steam to achieve efficient operation.  That is why nuclear plants have those huge iconic hyperbolic cooling towers, or are located adjacent to bodies of water into which they transfer heat.  The amount of heat they transfer can impact local ecosystems, not to mention the organisms destroyed in pumps and screens as they are sucked through the cooling system.  The US EPA and at least one state (California) are developing regulations to reduce or mitigate the impacts of this once through cooling process.
Another problem with nuclear power is what happens when things aren’t working the way they’re supposed to.  Because of the potential problems when something does go wrong, nuclear power plants are pretty much uninsurable.  Instead, governments legislate liability limits for nuclear plant owners or take responsibility beyond a certain level.  While this has been necessary to make investment in nukes commercially viable, it eliminates or at least mutes signals to engineer changes that would reduce the potential risks associated with something going wrong. 
What about the radioactive waste generated by these plants?  While the “preferred” solution of hauling spent nuclear fuel to a geologically stable location where it can be stored for the thousands of years needed for it to reach safe levels of radioactivity has not come to pass, dry cask storage systems make it possible for a nuclear plant to store all its spent fuel on site in a passively safe manner.  According to the World Nuclear Association[1], worldwide, there are about 270,000 tonnes of used nuclear fuel currently in storage with an additional 12,000 tonnes added annually.  Compare this to the 125 million tons of combustion by-products produced annually by coal power plants in the US.  Nuclear plants could actually store all their spent fuel on site for their entire operating life in containers that can be safely shipped to centralized storage or reprocessing facilities when and if they become available.

The Bottom line

Nuclear energy is an attractive base load generating resource that can produce large amounts of electricity without the pollution problems and global warming impact of plants that rely on combustion of fossil fuels.  Nukes require much less real estate than solar or wind generation and provide a much more predictable energy supply than these intermittent resources.  When the smaller scale, passively safe, factory built nuclear generators currently under development are licensed and become available, they may have an important role to play in our energy future.

Monday, March 21, 2011

Subsidizing Electricity Storage

One of the unique characteristics of the electricity grid is that it epitomizes the concept of just in time delivery.  The amount of electricity generated must exactly equal the amount consumed moment to moment.  Basically, whenever a light switch is turned on a generator somewhere has to increase its generation – ramp up – to provide the needed power.  That’s why we have all these fancy control rooms with computerized map boards like the one shown below.

That’s also why the prospect of non-dispatchable variable generators like wind and solar make system operators nervous.  Sudden shifts in the wind or moving clouds can cause rapid and unanticipated changes in generation which in turn requires other dispatchable generators to be available to increase or decrease their production to balance the changes blowing in the wind.  As the amount of variable generation increases, the potential magnitude of the balancing challenge increases with it.

One potential solution to this is to develop some kind of advance storage mechanism that can store excess generation and then release it when it is needed.  On one level there is nothing new here.  Some would argue that fast response gas turbines perform a storage function – storing ancient sunlight in the form of natural gas and releasing the energy in the form of electricity when needed.  Pumped storage hydroelectric facilities serve the same purpose – pumping surplus electricity up hill and then having it flow through turbines when needed to generate power.  Other technologies, like compressed air energy storage, can serve the same purpose.  But are they enough?  

Some advanced storage advocates argue that these battery or flywheel-based technologies are the answer because they can respond quickly and be built most anywhere.  Like photovoltaic panels, however, they need special treatment and subsidies to prime the pump and make them cost-effective.  Storage advocates have been successful in getting the California legislature to pass a low (AB 2514) requiring the CPUC to “Consider the Adoption of Procurement Targets for Viable and Cost-Effective Energy Storage Systems,” which has resulted in a rulemaking (R.10-12-007) to do just that.  Advocates claim that the fast turn around rates which allow for fast ramping overcome the energy limitations of these devices and somehow provide an improvement over gas turbines and available regulation resources that are currently used.  Others suggest a blatant attempt to get special treatment and subsidies for a “climatically correct” technology that would otherwise not be competitive.  The reasoned approach would be to identify renewable resource integration needs, specify ancillary services products to meet those needs, and then let the market decide which technologies do the best job of meeting the needs.  If fast ramping is needed and these new technologies are the best way to provide the service, there will be a demand without special carve outs or subsidies.

Monday, February 21, 2011

Competitive Transmission Providers?

One of the hot topics in California at the moment is the CAISO’s apparent favoring of incumbent utilities in approving new transmission projects.  Proponents of alternative transmission system developers claim that the ISO’s purported preference diminishes competition and ultimately increases cost to consumers.  They claim that they can build transmission facilities better, cheaper and faster than the big incumbent IOUs (Pacific Gas & Electric, Southern California Edison and San Diego Gas and Electric), and that the ISO is unfairly discriminating against the competitive developers just to keep the utilities happy.  If you look back at the development of non-utility generation and the merchant generation business, the competitors have a good point.  Over the last dozen years or so, the competitive market has put downward pressure on generation development and operating costs, shortening the development process, and shifting risk from the utility ratepayer to the developer.  This has been good for the generation business so surely it would benefit the transmission side of the business.

There is no doubt (except maybe in the “Southern” states and among the APPA) that the competitive generation business has been good for consumers, so we should strive to bring competition to other aspects of the business as well, right?  Maybe not.  The generation business clearly benefited from reduced costs and increased efficiencies brought about by competition.  However, one of the primary reason was not increased efficiency compared to bloated vertically integrated utilities (though it certainly was), but a completely different profit paradigm.  Merchant generations make their profits from selling energy at prices higher than their costs.  The most effective way to increase profits is thus to reduce costs and increase efficiency over both the short and long-term.  This encourages practices like hedging gas price risk and minimizing heat rate.  Utilities, on the other hand, were able to pass through “reasonable” expenses and recover a specified return on equity.  In other words, the more they invested in rate-base the more earnings they were able to return to their shareholders.  This peculiar “regulated cost of service ratemaking” was a function of the “natural monopoly” compact that was developed by Samuel Insull and implemented in the early 1900s. It was very effective for most of the century as electrification spread and marginal costs decreased.  But when competition is an option it makes little sense to reward companies for convincing regulators that they should build more stuff.

Unfortunately, that’s where we are in the transmission part of the business.  Because of the inter-connected and integrated nature of the transmission grid and the variety of impacts a transmission upgrade may have, it’s not reasonably feasible to charge for usage.  Also, the operating costs of a transmission line are trivial compared to the capital cost to build it, so that reduced operating costs have an insignificant impact.  Then there’s the fact that transmission covers a huge geographic area, making franchise agreements and access to eminent domain important characteristics.  So what exactly is it that “merchant” transmission developers have to offer that make them better suited to develop transmission projects?  Virtually all rely on the “go to FERC and get a guaranteed rate of return approved and have the ISO include the costs in its transmission rates” model, which bears a very strong resemblance to the utility model they’re proposing to replace.  Some might argue that it’s just a different set of shareholders.   

Thursday, February 10, 2011

Solar PV - Dis-economies of Scale?

Some recent announcements from Southern California Edison (SCE) appear to warrant more than a little head scratching.  SCE recently announced contracts for the purchase of energy from two different sets of sources.  On January 31, 2011, SCE filed the contracts it had announced in November for 239 MW (567 GWh/year) of solar PV projects resulting from its Renewable Standard Contract (RSC) program.  The 20 projects are all between 4.7 and 20 MW, are for 20 year terms, are scheduled to come on line between April 2013 and April 2014 and are all priced  below the 2009 Market Price Referent (MPR) ($108.98 for contracts starting in 2013 and $112.86/MWh for contracts starting in 2014).  This would certainly appear to support assertions that PV prices are coming down dramatically and could soon be competitive without massive subsidies. 
Also in January, SCE announced that it had executed seven contracts totaling 831 MW for solar PV resources between 20 and 325 MW each, coming on line from 2103 through 2016.  These contracts are all priced ABOVE the very same MPRs that the smaller RSC contracts are below.  How could it be that smaller projects using the same technology are less costly on a per unit basis than larger projects?  I’ve no idea, and not being privy to the confidential pricing terms of the contracts, I can only guess.
The first issue is one of scale – PV installations in the multi-MW size range all pretty much use the same panels, inverters, transformers and other equipment.  A larger project that interconnects at transmission voltage (115-230 kV) will require an extra step of transformation and more costly interconnection facilities than a smaller project that interconnects at distribution voltage (12-33 kV).  Review of SCE’s filing of the RSC projects does show that they are almost all interconnected at distribution voltage.  Since distribution facilities cannot handle anything much larger than 20 MW (if that), larger installations would have a “dis-economy” of scale based on interconnection voltage.
Another issue relates to parcel size and permitting.  At roughly 10 acres per MW, a 200 MW project requires a huge tract of land – over three square miles!  Such a project is likely to get much more attention in the permitting process than a 10 MW project that “only” needs about 100 acres.  It is also more likely to have a significant environmental impact and require both more complex permitting and greater mitigation costs.  Here again size is a dis-economy.
Then there is the cost of capital.  A 200 MW project will cost upwards of a billion dollars.  Raising that much capital probably incurs a higher cost and may also require a higher percentage of equity to secure debt financing.  Both would tend to increase the unit cost of the project compared to smaller projects.
Then there is the PPA negotiation process.  SCE developed the RSC (and the CPUC approved the related Renewable Auction Mechanism, aka RAM) as standardized contract with limited room for negotiating terms and conditions.  Bidders are required to bid on price alone head-to-head against other potential projects for a contract.  They are likely to offer their lowest and best price into the solicitation. Large RPS contracts, on the other hand, often require months of negotiation and in many cases go through upward pricing adjustments before they come to fruition.  How much this adds to the final price is hard to say.
The conclusion drawn from these interesting submissions is that an increased reliance on distributed PV generation – whether on rooftops, vacant lots, or surplus agricultural land – may not prove to be more costly than the huge projects built in the middle of nowhere.  Of course, the smaller projects do not require the massive transmission projects needed to export power from the middle of nowhere.  That means less utility ratebase – often receiving incentive rates of return – available to benefit IOU shareholders, the very same IOUs that are negotiating PPAs with the massive projects.  Interesting coincidence.