Thanks to Google's master plan to digitize every word in print, I recently found an online copy of a February 1883 edition of The Electrician that reprinted part of a Thomas Edison interview on the subject of stationary energy storage. For readers who know that history doesn't repeat itself but it frequently rhymes, I've posted an HTML version of the article as an Instablog.
Since The Electrician article was the source for several infamous Edison quotes, I was thrilled at the opportunity to better understand why Mr. Edison would say things like, "The storage battery is one of those peculiar things which appeal to the imagination, and no more perfect thing could be desired by stock swindlers than that very self-same thing," and, "Just as soon as a man gets working on the secondary battery it brings out his latent capacity for lying." I couldn't help but chuckle over a note that the editors, "left out some of the more strongly worded sentences."
Reading the entire article was a short-course in humility that showed me once again, even careful research can lead to a wrong conclusion if you don't drill down far enough into the detail. When I first looked into the early history of electric lighting, I assumed Mr. Edison was an angry customer who'd been hoodwinked by a battery company. Instead of finding what I expected, I learned Mr. Edison was simply upset that his less than honorable contemporaries were selling fantasies to investors while destroying public confidence in the emerging electric light industry.
The scam worked like this. A promotional company that claimed to have proprietary power and lighting technology sold the 19th Century equivalent of franchises to investment syndicates that were organized for the purpose of providing electric light in a particular territory. The promoters charged up-front fees for license rights and then sold the syndicates all necessary equipment and materiel. While the syndicates were dismal failures, the promoters made a fortune and apparently raised the current equivalent of $700 million in the UK before migrating to the US with a refined scheme that included an insidious new twist - In addition to selling license rights, generators and materiel, the promoters added batteries to the mix so that syndicates could run their generators during the day to charge batteries in their customers' homes, and run their generators at night to power even more electric lights.
Mr. Edison's objections weren't surprising. He started by explaining that he patented the idea of combining generators, batteries and electric lights in 1879, but batteries were the weak link in his plan. After explaining the technical challenges, Edison offered this blunt economic assessment, "Scientifically, storage is alright, but, commercially, as absolute a failure as one can imagine." Using batteries to store electrons for their energy value was simply a money-losing proposition.
A hundred and thirty years later, the idea has come full circle and stationary energy storage is entering a new epoch where opportunities seem larger than life and challenges are larger than they seem. In mid-December the DOE submitted a report to the Senate's Energy and Natural Resources Committee titled "Grid Energy Storage" that details the benefits of grid-scale energy storage, summarizes the economic and technical challenges and discusses the ongoing efforts of government, academia and industry to meet those challenges.
The principal policy driver behind the DOE's plan to implement grid-based storage at massive scale is increased reliance on intermittent power from renewable resources like wind and solar. While the report gives lip service to other goals like improving resiliency, enhancing reliability and facilitating emergency preparedness, it's really all about mitigating pollution of the electric grid with intermittent renewable power.
This "Duck Curve" from a "Demand Response and Energy Efficiency Roadmap" that CAISO published in December highlights the problems by showing the historic, current and projected impact of photovoltaic power on its net system load for a typical spring day when PV output is high and air conditioning loads are low. To help readers follow my reasoning on what the Duck Curve shows, I've added a box labeled "PV Prime Time" to accentuate the six-hour period when solar panels can be more than trendy roof-mounted status symbols.
While CAISO has not published a comparable graph for a typical summer day, the red line in the following graph from a recent PG&E presentation on the impact of demand response programs shows net load for a summer day when electricity use hits its annual maximum. Once again, I've highlighted PV Prime Time with a yellow box.
When I saw the Duck Curve my first reaction was surprise that CAISO's 2013 system loads in PV Prime Time were about the same as its 3 am loads. Where is all that surplus off-peak power ideologues use to promote electric vehicles? I certainly can't see any in California.
My second reaction was surprise that the Duck Curve shows why PV solar can reduce fuel use in power plants, but it can't reduce the overhead costs of building and maintaining those plants in the first place. PV Prime Time is about six hours out of phase with actual demand. The perverse outcome of widespread PV deployment is that huge capital investments in PV solar result in under-utilized conventional resources, which increases the unit cost of all conventional power.
My third reaction was skepticism over the smooth load curve during PV Prime Time. Everybody understands that PV Solar is almost as reliable as a part-time employee with a drug habit. Solar output can drop by 80% with a passing cloud and a rainy day can eliminate the solar belly. Once again, the perverse outcome is higher utility overhead for frequency regulation and spinning or standby reserves to ensure that aggregate demand is met regardless of the weather.
My final reaction was more nuanced but critically important. The Duck Curve shows that "net metering" encourages PV system owners to produce power during the solar belly and offset that relatively low value power contribution against the relatively high value power they use in the evening peak. This reality is diametrically opposed to prevailing mythology that net metering creates power when it's needed most and the offsetting benefits are derived in off-peak hours.
At this point a sensible person might ask, "Why in heavens name does anyone believe residential PV solar is a good idea?" The answer is complex, but it boils down to poor customer education, ideologically motivated regulation, the politics of electric power and big differences in the way utilities bill residential and commercial customers.
In a publication for commercial ratepayers titled, "Understanding Electric Demand," National Grid explains that the cost of electric power includes three distinct elements:
- The actual cost of generating electric power;
- The cost of owning and maintaining distribution networks; and
- The cost of operating overhead and profit.
It further explains that there's relatively little variation in electricity use from house to house, but "this is not the case among commercial and industrial energy users, whose electricity use - both consumption and demand - vary greatly. Some need large amounts of electricity once in a while - others, almost constantly." Because of these differences in consumption patterns, residential customers pay a combined charge based on total energy use while commercial and industrial customers pay a "demand charge" for infrastructure and an "energy charge" for the cost of power generation.
If you think about residential PV solar for a minute, it undermines the rationale for combined charge billing. For six hours a day, system owners are power producers who want their utility to accept excess power and provide redistribution services. For eighteen hours a day, they're power consumers who expect their utility to provide an offset for six hours of excess power and then provide additional power for the same price other consumers pay.
The following discussion summarizes PG&E's basic residential and commercial tariffs to give you a sense of what the power cost breakdown looks like.
During summer months, a typical "Tier 1 Residential Customer" with time of use billing under the Schedule E-7 tariff pays:
- A meter charge of $3 a month,
- An energy charge of $0.55 per kWh for power used between 12 noon and 6 pm, and
- An energy charge $0.31 per kWh for "off-peak" power.
While the Schedule E-7 tariff provides lower rates for low-income customers that use from 100% to 201% of its baseline, those customers aren't the target market for residential PV solar.
In contrast, a commercial customer with "secondary" electricity service of less than 2,000 volts and time of use billing under the Schedule A-10 tariff pays:
- A meter charge of $140 per month,
- A demand charge of $13.36 per kW for their highest 15-minute power use interval; and
- An energy charge of $0.165 per kWh for power used between 12 noon and 6 pm, and
- An energy charge $0.135 per kWh for "off-peak" power.
The injustice of net metering is that PV Solar systems are economic assets that create immense variability in the power consumption of residential customers who expect their utility to offset their relatively low-value contribution against their highest-cost demand. The end result is that taxpayer subsidized purchases of PV solar systems increase power costs and decrease power quality for everybody else while giving PV system owners a warm fuzzy glow because they're taking affirmative action to help make the big blue marble a little greener. Ain't life grand?
Over time I think utilities will move away from net metering and combined rate billing for residential customers who decide they want to be part-time power producers. When that happens PV system owners will be paid a fair price for the power they generate and charged a fair price for the power they use. They may also find themselves paying demand charge analogs to reflect the reality that producing a little solar power in the afternoon is not a fair offset against peak evening demand.
As a long-term supporter of ideologically motivated and economically questionable alternative energy schemes, the DOE is facing a monstrous problem. It either has to find a cost-effective way to use renewable power to satisfy peak demand, or admit that intermittent renewable power can't possibly satisfy the basic demands of an industrialized economy where consumers expect the lights to work when they flip a switch. Since the latter is politically unpalatable, the DOE has embarked on a quixotic quest for the Holy Grail of Cleantech - cost effective energy storage.
In describing overriding policy goals, the DOE's report to the Senate explained "The future for energy storage in the U.S. should address the following issues: energy storage technologies should be cost competitive (unsubsidized) with other technologies providing similar services; energy storage should be recognized for its value in providing multiple benefits simultaneously; and ultimately, storage technology should seamlessly integrate with existing systems and sub-systems leading to its ubiquitous deployment." That's a tall order, particularly when the phrase "other technologies providing similar services" means gas fired peaking plants and pumped hydroelectric storage, a wonderful technology with limited potential for expansion.
The report also states that with the exception of pumped hydro, current storage technologies are not cost-effective when it comes to shifting power from the solar belly to peak evening demand. While the language is more than a little oblique, the DOE's core conclusion is that the world needs better, cheaper and more durable batteries, but without step-change advances in battery technology, using them to store electrons for their energy value will be a money-losing proposition anywhere that has a stable power grid that doesn't rely on oil-fired generation.
The biggest challenge the energy storage industry faces is economics. Using PG&E's residential tariff the maximum value of moving a kilowatt-hour of electricity from off-peak to peak is $0.24, or $87.60 per year. Since the unsubsidized price of stationary storage systems SolarCity (NASDAQ:SCTY) is offering in cooperation with Tesla Motors (NASDAQ:TSLA) is about $1,500 per kWh, the fundamental economics are dreadful. The only way the economics can work for end-users is a combination of government subsidies, utility incentives and pre-packaged financing programs that absorb 100% of the cost savings but insulate owners from future electric rate increases.
In its report to the Senate, the DOE noted, "the storage component still constitutes only 30% to 40% of the total system cost." The other 60% to 70% of total system cost is attributable to:
- Battery module and pack assembly;
- Temperature management and safety systems;
- Battery management systems;
- Power control and conversion systems;
- Integration, engineering and installation;
- Marketing and distribution expenses; and
- Profit for each participant in the value chain
While I won't belabor the point, Tesla reportedly pays about $250 per kWh for the lithium-ion battery cells it's building into stationary energy storage systems that cost end-users $1,500 per kWh. While cell costs are and will continue to be an important part of the mix, there can't be meaningful progress on total system costs without slashing the "balance of system costs" and the costs of deploying storage in end-user facilities.
For the last decade lithium-ion batteries have enthralled ideologues, dreamers and government because energy density is a mission critical requirement for batteries in electric cars. They've argued that lithium-ion technology is a silver bullet for all energy storage needs and that demand from automotive would drive battery prices down to a level where lithium-ion would be a natural choice for stationary systems as well. The facts that were almost universally overlooked are:
- Lithium-ion battery technology and architecture reflect decades of refinement and optimization by some of the world's best manufacturers. The idea that fine companies like Panasonic, Sony, LG Chem and Samsung have overlooked opportunities for major product improvements and cost savings is preposterous;
- Energy density is a meaningless metric in stationary systems because the size and weight of a shipping container of batteries in a parking lot doesn't matter; and
- The unique safety and performance profiles of lithium-ion batteries significantly increase the balance of system costs that represent the lion's share of the end-user price.
There will be performance gains and cost reductions as innovations are developed, validated and integrated into the world's battery manufacturing infrastructure. But those gains will be modest and the implementation timelines will be measured in decades.
Several years ago I spent an afternoon talking with the president of Enersys (NYSE:ENS) Europe about their customer experience in the stationary and motive markets. He explained that new customers always walked through the door thinking they needed a lithium-ion solution, but after comparing their application requirements with the available technology options and considering their all-in cost for a fully integrated system, the vast majority of customers chose lead-acid because it was a better fit for their technical needs and their budgets.
The bottom line is that companies and communities are happy to spend somebody else's money on high profile lithium-ion battery demonstration projects, but when it comes to spending their own money on stationary energy storage they're more likely to buy from established energy storage technology providers like Enersys and emerging technology developers like Axion Power International (NASDAQ:AXPW) and ZBB Energy (ZBB) that have been largely ignored by the media and government, but offer better total cost of ownership profiles because of their lower balance of system costs.
Since November 2008 Enersys' stock price has steadily climbed from a post-crash low of $5.71 to a current price of $69.33. It nailed a twelve-bagger without government support, without flashy projects and without hype. The gains were attributable to solid execution and staying the course while less committed battery manufacturers lost their way. During the same period Axion and ZBB have both stayed on course and made tremendous technical strides despite stock charts from hell. As the stationary markets develop and pending projects turn into firm orders, I expect stellar performance from both companies.
In 1910, Thomas Edison observed, "When we learn how to store electricity, we will cease being apes ourselves; until then we are tailless orangutans." Cost-effective energy storage presents immense challenges, but there are solutions and new technologies are certain to emerge in time. Ultimately the question for investors is "Would you rather make headlines or money?" The hype machine has kicked into overdrive on a pipe dream of PV solar combined with lithium-ion batteries. The reality is that boring technologies with better price-performance are the lead-pipe cinch because the laws of economic gravity cannot be ignored.
I want to thank Batteries International Magazine and cartoonist Jan Darasz for giving me the chance to share this cartoon that speaks volumes about the battery and solar power industries with readers.
Disclosure: I am long AXPW. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: The author is a former director of Axion Power International (AXPW) and holds a substantial long position in its common stock.