Exclusively to GTM, Assistant Los Angeles County Supervisor Norm Hickling confirmed rumors of “a large layoff” of the AVSR1 workforce on Friday. “I also understand negotiations continue,” Hickling added about attempts to settle the dispute between First Solar and LA County that has put the project six weeks behind schedule, with “little progress.”

Approximately 120 of the current 240 person workforce was “furloughed,” according to First Solar Public Relations Director Alan Bernheimer.

This is a severe blow to an already jobs-poor Los Angeles-adjacent community where recent studies found as many as one in three homeowners behind and/or underwater with their mortgages.

Negotiations began in early April after an LA County safety inspector discovered that electrical connections on the 3.7 million First Solar cadmium telluride (CdTe) thin film photovoltaic panels that were to be used at the site did not have the Underwriters Laboratory (UL) approval required by the project’s County-granted Conditional Use Permit (CUP).

The inspector was reportedly doing a routine site visit, in service to the CUP, when he discovered there was no UL approval for the panels’ electrical connections.

State of California building and safety codes require UL approval of electrical connectors. That approval is apparently not required or its absence was not noticed in other states, such as Nevada, where First Solar has done thin film installations. It is also a standard to which First Solar has not been held at installations it has done elsewhere in California.

Neither the County nor First Solar have offered many details.  “The Los Angeles County Public Works Department,” County spokesperson Bob Spencer told GTM when negotiations began, “is working with First Solar to address its plan check comments relative to the rating of the modules and the applicable electrical safety regulations.”

“First Solar is in discussions with Los Angeles County Public Works regarding electrical codes and standards interpretations as they relate to utility-scale, solar photovoltaic (PV) installations,” First Solar spokesperson Ted Meyer told GTM at that time. “We are confident these discussions will result in a satisfactory resolution.”

That was six weeks ago.

“Our discussions with the county are ongoing and we are working to resolve the issue so we can put people back to work,” Bernheimer told GTM Saturday.

First Solar spokesman Adam Eventov told the Oso Town Council on Thursday night a settlement was expected by mid-June but noted, in the same statement, that problems with dust were preventing the company from continuing to grade land and prepare racks for panel installation at the project site. A proposal to use decomposed granite to minimize dust had been blocked by the County, Eventov said, leaving them with only the expensive alternative of hydroseeding.

Eventov also said that First Solar has, as promised, made a $140,000 donation to Antelope Valley College on behalf of the AVSR1 project but continues to hold a matching donation to the local communities until panel installation can continue.

At the same meeting, elected members of the Oso Town Council Richard Skaggs and Gerry Conroy, just back from the annual First Solar shareholder meeting in Phoenix, reported to the community that they thought the County was more responsible for impediments to continued construction than the company.

Panel installation was scheduled to begin at AVSR1 by mid-April. If the panels cannot be approved until UL certification is obtained or replacement with another technology becomes necessary, the financial consequences for First Solar could be problematic to the already financially hamstrung company. First Solar’s share price was as high as $140 in July 2011 but now hovers around $15.

First Solar continues to be in charge of engineering, procurement and construction (EPC) at the site. It is also in charge of EPC at the NRG Energy Alpine Solar project located a few miles away which, like AVSR1, would use the problematic panels and is subject to provisions of an LA County CUP.

The total number of panels that will require some kind of attention or adjustment is well in excess of four million.

Neither the County nor First Solar would comment on whether corrections at other First Solar projects in development or retroactive corrections of panels at other California sites will be necessary.

Also under development with First Solar as panel supplier and in the EPC role and -- therefore potentially affected by this issue -- are the 550-megawatt Topaz Solar Farm in San Luis Obispo County owned by MidAmerican Holdings and the 550-megawatt Desert Center Solar Farm in Riverside County owned by NextEra Energy and GE.

The 21-megawatt Blythe Solar Project in Riverside County owned by NRG Energy may also be affected.

With a power supply uncertainty looming due to the safety issues that took the 2,200 megawatt Son Onofre Nuclear Generating Station (SONGS) off line, it would have been a positive result for Southern California if AVSR1 was coming online this summer.

In today's podcast, GTM Solar Analyst team members, Shyam Mehta and Shayle Kann discuss the gory details of the ruling's impact. Scott Clavenna, Greentech Media's CEO, leads the discussion. We also cover the recent run of funding in solar and give thanks and props to some of GTM's many article commenters.  

We look forward to providing these podcasts as a regular feature and bringing you the events of the week in a different format. You'll hear from GTM research analysts, editors, reporters and the occasional special guest. Stay tuned.

Since early this spring, Little has been predicting a tariff in the 30 percent range would be put on Chinese solar panel imports. On May 17, the U.S. Department of Commerce imposed a 31 percent tariff on panels from China’s biggest panel makers.

U.S. solar manufacturers have been struggling against low-price panels from China. The tariff is intended to prevent what domestic trade groups have described as “dumping” by the Chinese of below-production-cost products. Chinese panel makers are attempting to unload excess inventory even though they are being forced to do so at a loss.

There are two parts to the Commerce Department tariff, one punitive and one preventative. The 61 Chinese companies identified with the dumping will be required to pay the 31 percent tariff. To prevent those companies from shipping into the U.S. under a false front, Chinese panel makers who have not been exporting to the U.S. will be required to pay a 250 percent tariff if they begin doing so.

Some say Chinese manufacturers are likely to get around the tariffs by using a non-Chinese false front. Others fear China will counter-accuse U.S. polysilicon manufacturers of unfair practices before the World Trade Organization.

Media reports say Chinese solar company stock prices fell with the Commerce Department announcement while U.S. panel makers’ share prices rose.

“From a broad view,” Little recently told GTM, referencing data that put the U.S. imported price of a Chinese module at 94 cents per watt, “a tariff would not be good for the industry because it would likely result in higher-priced modules.” But, as with the cost of Spire’s turnkey facilities, Little said, variables make all the difference.

In the same assessment from which he got the 94 cents per watt Chinese module price, Little found that U.S. manufacturers would require a 31 percent tariff if they were to import Chinese wafers to manufacture modules. The wafer price would be 35 cents, the wafer to module conversion cost would be 88 cents and the U.S.-made module would cost $1.23.

Domestic manufacturers would, however, only need a 14 percent tariff if they were to import Chinese solar cells at 53 cents to make modules. The cell-to-module conversion would cost 54 cents, making the cost of a U.S. module $1.07.

And, Little found, if a U.S. manufacturer were to import laminates at 70 cents, a laminate-to-module conversion cost of 24 cents would make the price of a U.S. module cost 94 cents and no tariff would be needed.

Spire offers turnkey assembly lines that perform all three of those manufacturing processes.

Spire, Little said, will continue to thrive despite the tariff. They have customers all over the world. They rode the Chinese expansion of the last few years very successfully and have turned their attentions more recently to India, where a 20,000-megawatt solar target, incentive programs and an entrepreneurial inclination are sparking a major expansion. His company has also recently done business in such disparate and expanding markets as Ethiopia, Turkey and Romania.

“We know what’s going on all over the world,” Little acknowledged. He had no fear of a backlash or a trade war from the tariff but, in fact, saw it as an opportunity.

Germany-based SolarWorld and Japan-based Sharp, which according to Little dominate the U.S. crystalline silicon PV market, obtain their manufacturing equipment in their home markets. Spire, Little said, therefore has no chance to win their business.

A tariff, though, will bring new customers. “We talk to the top ten Chinese manufacturers,” Little said. “We have a come-to-America program. We’ll give you the machines, we’ll set you up, we’ll help you do the whole thing in America.”

Ultimately, Little said, the real problem will come if the tariff impedes the advance of the levelized cost of solar energy-generated electricity in the U.S. and around the world toward grid parity.

Module prices may rise, Little predicted, but a healthy competition will return to the broader solar sector and that could drive prices down. “I would’ve loved to have seen the concentrated solar cell (CPV) market take off. But CPV has the same problem everyone else has,” he explained. “There’s no traction because of these 88-cent modules coming in.”

Parity, Little said, is “a systems-driven equation.” The total calculation could remain competitive, he insisted. “If you start with the laminate, you can compete right now.” But, he said, if you want to start with cells, it’s going to cost you a little more.”

To keep the price low, U.S. manufacturers have to learn where the competitive point to enter the value chain is, Little said. “That’s the whole point.”

The tariff will also shift the international market, Little acknowledged. Chinese laminate manufacturers will find a ready market in the U.S. but cell and wafer manufacturers “will have a difficult time being competitive” unless the tariff drives them “to manufacture in the U.S.” Likewise, he added, with the knowledge of where to enter the values chain, domestic manufacturers “can compete then, too.”

With the unofficial start of summer just around the corner, utilities and independent system operators are looking for ways to relieve peak loads and congestion on the electricity grid for the coming season and down the road.

Although every region of the country claims it has adequate generation for even the hottest days of summer, each ISO is also looking forward as U.S. Environmental Protection Agency rules will take affect in coming years.

Grid operators are planning now for how to replace coal with natural gas, along with a side of demand response and a dash of renewables. Many of the changes won’t come for years, but the planning is happening now and changes to pricing and rules, will happen ahead of the EPA rules.

“From a PJM perspective, the lights aren’t going to go out,” Paul Sotkiewicz, chief economist
 at PJM Interconnection, said during a Restructuring Today webinar on Wednesday.

There are two EPA rules that are putting pressure on older coal generation, the cross-state air pollution rule, which is currently on stay until further ruling, and the MATS rule, which regulates mercury and other heavy metals and is scheduled to go into effect in April 2015.

PJM Invests in Transmission

In PJM’s territory, the problem is not a lack of capacity. Sotkiewicz said that there’s quite a bit of natural gas that isn’t being used to its full capacity. PJM expects that nearly 14,000 megawatts of coal-fired generation will retire by 2016. Most of the plants are over 40 years old or 400 megawatts or less, which makes up about 30 percent of PJM’s current coal fleet.

In anticipation of the changes, the capacity auctions for the 2015 to 2016 year in PJM territory procured a record amount of new generation for one year, nearly 5,000 megawatts. Almost all of the new generation is gas-fired. For the first time ever, new combined-cycle gas plants are the same price as a northern Appalachian coal plant, according to Joseph Bowring, independent market monitor for PJM and president of Monitoring Analytics.

Overall, PJM has 164,561 megawatts of capacity resources. Along with new gas resources, the capacity auction procured nearly 15,000 megawatts of demand response and energy efficiency. Energy efficiency, solar and wind saw double-digit growth since the year before, although the renewables are still just a drop in the bucket in terms of generation. Solar, for example, increased to 56 megawatts, a 22 percent increase from the year before.

To combat retiring coal generation, PJM recently announced $2 billion in transmission upgrades. There will be more than 130 projects that range from substation upgrades to rebuilding transmission lines. More than half the projects will be in Ohio, where there is a lot of coal retiring. However, the upgrades could also be a boon to renewables like wind, which has various transmission constraints.

MISO Seeks More Wind, Demand Response

The Midwest Independent System Operator is facing essentially the same problem as PJM, although it is even more pronounced, since more than 70 percent of its fleet is coal-fired. MISO also has transmission constraints that are being revisited, but there have been no announcements about upgrades similar to what PJM is doing.

The constraints will mean a change to demand response rules, said Jameson Smith, manager of regulatory studies at MISO. “Demand response qualification requirements and deployment procedures will be reviewed and given a potential increase in use.”

For MISO, there’s more wind that’s currently slated to be built than gas, but that could also change in coming years, said Smith. Any new gas could need new pipeline infrastructure, which can take at least five years to develop. At this time, “there’s limited ability to increase existing gas-fired unit capacity,” said Smith. But wind needs transmission too, which means that no matter what comes on-line to replace coal in the next five years, there will have to be substantial investment.

ERCOT Pushes Up Prices

In Texas, ERCOT is already planning to spur new generation by raising the system-wide offer cap. Last summer, prices hit $2,500 per megawatt-hour as ERCOT’s grid was strained by high temperatures. If a ruling now being considered is approved, that could jump to at least $4,000 as early as this August, and perhaps rise considerably more in coming years. “There’s a consensus that the system-wide offer cap will have to go significantly higher,” Ken Anderson, commissioner of the public utility commission of Texas, said during a Restructuring Today webinar. “We need to at least double it -- and maybe triple it.”

A bump to the SWOC will also come with a change to a variety of other rules, including a potential increase of 200 megawatts to come from energy efficiency programs and a decision that could allow waste gas and energy storage to play in ERCOT’s markets. The cost of a negawatt is still considerably cheaper than building new generation. 

Anderson said that he believed that the commission was removing most of the barriers to energy storage coming into the wholesale market, and now it was just a matter of economics. The next few years will also be rosier for negawatts. “Eventually, I’d like to get demand response out of energy efficiency completely and into the competitive space,” he said during the webinar. “We need to give some certainty to where the market will be in 2014, 2015 and beyond.”

A fight over the future of net energy metering (NEM) in California was resolved by a  California Public Utilities Commission (CPUC) May 24 decision on the arcane question of how to define the NEM cap. The definition of the cap had become a battleground over NEM pitting Investor Owned Utilities (IOUs) against renewables advocates.

Following speeches in which they noted the many economic benefits to California from renewables, the commissioners voted unanimously against the IOUs and in favor of a definition of the NEM cap that will allow for much more distributed generation (DG) going forward.

“This is a big, big win,” said Mainstream Energy Director of Government Affairs Ben Higgins.

Like 43 other states, California has a NEM program that allows owners of DG systems of up to one megawatt in capacity, like small wind turbines, combined heat and power systems and rooftop solar systems, to reduce their electricity bills.

For the kilowatt-hours they send to the grid, system owners’ meters turn backwards as they are credited at the same retail rate they pay for the kilowatt-hours they consume.

When California established its NEM program in 1995, it imposed a 0.1 percent cap but used the ambiguous language of “aggregate customer peak demand” to define what the total megawatts of net metered systems should be divided by to calculate the cap percentage. And that calculation remained undefined, even as the CPUC expanded the cap to today’s five percent.

The differing methods used by the IOUs to calculate the bottom term of the cap equation, and the differing percentages thereby obtained, were recently observed by the Interstate Renewable Energy Council (IREC) which, among its other activities, acts as a watchdog group on U.S. net metering programs. IREC filed a motion asking the CPUC for clarity. Commission President Michael Peevey issued a proposed decision April 5.

He pointed out several differences in how the IOUs calculate the percentage of their NEM but noted one key commonality: Pacific Gas and Electric (PG&E), Southern California Edison (SCE) and San Diego Gas and Electric (SDG&E) all use “coincident” peak demand. Renewables advocates argue that “non-coincident” peak demand should be used.

Coincident peak demand is the designated period when all sectors (residential, commercial and industrial) reach their maximum electricity consumption and the state’s consumption peaks.

Non-coincident peak demand is the sum of the individual peaking demands of all customers in the three sectors. Residential peak is typically late afternoon, commercial peak is early midafternoon, and industrial peak can be at night. That sum of all peaks is greater than the total peak demand at any one time of the day.

When the installed DG capacity eligible for NEM divided by the peak demand gets to five percent, the utilities are off the hook. So they want that bottom number to be smaller. Renewables advocates want just the opposite because the larger number keeps what one solar advocate called their “backbone” incentive in place.

Peevey concluded that the legislature “did not intend ‘aggregate customer peak demand’ to mean coincident peak demand…[and] SCE, SDG&E, and PG&E should use the aggregation of customers’ non-coincident peak demands to calculate their caps on NEM participation…” The commission voted 5-0 to validate Peevey's decision.

In comments filed by their attorneys, the IOUs disputed Peevey’s conclusions. PG&E’s filing complicated the basic dispute by suggesting a change in the way both numbers would be calculated and concluded, “PG&E recognizes that this means more net metering. However…[it] is the better measure of the impact on the grid…”

By raising the issue of the impact of renewables on the grid, PG&E exposed the heart of the real debate between renewables advocates and the IOUs.

The utilities pointed out that of the three parts of the standard electricity bill, only one covers the price of electricity generated. The other charges cover the costs of delivering electricity through the transmission and distribution infrastructure. When NEM customers’ bills are reduced by the retail rate, they escape paying their fair share of costs for infrastructure they use as much as non-NEM customers. And, the utilities argued, it shifts costs to other ratepayers.

But the difference between the generation cost and the full retail cost of electricity is not necessarily a subsidy if the cost shifted to other ratepayers pays for benefits to them as well.

Consulting firm Crossborder Energy principal Tom Beach did a thorough cost-benefit analysis that was based on PG&E data and included a review of two previous cost-benefit analyses. It showed that if the higher value of the power not consumed due to the use of DG is considered, the benefits to the utility are greater and the costs to the other ratepayers are offset.

The biggest component of the benefit, Beach said, is the savings on power plant use and fossil fuel use. Such energy and capacity savings, Beach said, comprise 60 to 70 percent of the benefit to all ratepayers from the rooftop solar facilitated by NEM. And, Beach added, transmission and distribution system savings avoid the costs of line losses and the need for new transmission that provide another ten percent to twenty percent of the benefits from NEM.

Beach’s calculations came to a net benefit from NEM of two cents per kilowatt-hour for commercial and industrial systems, a cost of two cents per kilowatt-hour for residential systems, and, in sum, no cost extra cost of any significance to ratepayers.

We're now adding MiaSolé's 15.5 percent efficiency mark for a flexible CIGS solar cell. Note that this is an aperture-area efficiency on a commercial-size flexible PV module with a total area of 1.68 square meters. That eclipses the 13.4 percent mark recently set by SoloPower. MiaSolé is targeting commercial deliveries of 14-percent-efficient glass modules by the end of the year.

MiaSolé placed third in CIGS panel production in 2011, behind Solar Frontier (at 400 megawatts) and Solibro (at 66 megawatts), according to GTM Research. The firm also recently announced a 17.3-percent-efficient champion device, while the "manufacturing process for 14 percent efficiency is now in production." The firm recently made a rare presentation in Palo Alto, California to the Silicon Valley IEEE PV Chapter.

Some of the following milestones represent 'hero experiments,' but nevertheless -- the numbers keep rising. Here are some recent announcements of record-setting results:

Heliatek set a record for organic solar cells. It's a champion cell on a small area, but it has achieved 10.7 percent efficiency. The efficiency value for the 10.7 percent champion cell would be about 9.0 percent when deposited on a flexible substrate. The question remains: can organic solar cell technology be successfully commercialized in an unforgiving solar market dominated by crystalline silicon and First Solar?  Back in late 2009, Heliatek raised $27 million to build its first factory from venture capital investors Wellington Partners, RWE Innogy Ventures, and BASF Venture Capital, as well as industrial giant Bosch. 

SoloPower now boasts an NREL-measured aperture area efficiency of 13.4 percent. Module efficiency is significantly less than that. The value proposition for flexible modules from SoloPower and others is that there is less hardware required to install and the installation is easier. This thesis has yet to be proven in volume and scale. SoloPower builds flexible solar panels in a roll-to-roll electroplating process.

Suntech's (NYSE: STP) Pluto cell technology achieved a 20.3 percent efficiency for a production cell using commercial-grade p-type silicon wafers. Pluto technology is a combination of different elements which are brought together to improve cell efficiency, with 21 percent efficiency targeted within the next year. These incremental improvements include surface patterning, improved metallization, improved front metal contact dimensions, changes in dopant concentration at the emitter, and improved high-temperature performance. None of these processes come cheap. Plus, the new product has not exactly replaced Suntech's existing lines -- it appears to remain a premium product that is offered at premium prices.

Solar Frontier is number two in thin-film solar and number one in the CIGS/CIS race, with 400 megawatts shipped in 2011. The firm just racked up a 17.8 percent aperture-area efficiency on a 30-centimeter-square CIS-based PV lab module. The result was claimed to come on a "fully integrated submodule" performed with processes "very similar to what is in place" in Solar Frontier's factories at commercial production scale, according to a release from the firm. The Japanese firm's Kunitomi factory recently built a champion module at 14.5 percent aperture efficiency, equivalent to a 13.3 percent module efficiency.

First Solar (NASDAQ: FSLR) hit a new world record for CdTe PV module efficiency with a 14.4 percent total area efficiency in January. That mark comes six months after First Solar hit a CdTe solar cell efficiency of 17.3 percent. Both records were set at the firm's Perrysburg, Ohio factory.

Alta Devices' most recent gallium arsenide (GaAs)-based solar panel boasts a 23.5 percent efficiency, as verified by the National Renewable Energy Laboratory (NREL). The firm claims that "this is the highest solar panel efficiency yet achieved." The press release did not discuss the size of the panel and the company has not yet responded to our inquiry.

Alta Devices has won more than $120 million in venture funding from August Capital, Kleiner Perkins Caufield and Byers, Crosslink Capital, DAG Ventures, NEA, Presidio Ventures, Technology Partners, Dow Chemical, AIMCo, Good Energies, Energy Technology Ventures, and Constellation Energy. The firm is still in the pilot manufacturing phase. Chris Norris, the CEO of Alta, has said that the company's goal is to "compete with fossil fuels without government subsidies" and get to a levelized cost of energy of $0.06 to $0.07 per kilowatt-hour. The epitaxial lift-off technique pioneered by Alta founder Eli Yablonovitch allows the firm to produce layers of GaAs that are flexible and measure only one micron in thickness.
 

SunPower has been the heavyweight champion of the world when it comes to commercialized cell and module efficiencies for the last half-decade -- and by a significant measure. The company's back-contact crystalline silicon cell design, in commercial production since 2005, moves the metal contacts to the back of the wafer, maximizes the working cell area, and eliminates redundant wires. SunPower has been able to achieve consistent improvements in efficiency with each successive generation of commercialized cells, and this has translated to gains in the module arena, as well. The firm's Gen 3 cells have efficiencies in excess of 23 percent.

Solar Junction, a developer of multi-junction cells for high-concentration photovoltaic (HCPV) applications, is working with Semprius and has inked an agreement to deliver multi-megawatts of epitaxial wafers. Semprius recorded a module efficiency of 33.9 percent.

Abound Solar, a manufacturer of cadmium telluride PV modules, announced the production of 82.8-watt modules at its Longmont, Colorado factory, representing a 12.2 percent aperture efficiency that is now being verified by NREL. The units were produced on "existing production equipment," according to the firm's press release. Abound claims to have produced its millionth module in December 2011.

“We’ve got 1,700 [people] at work,” said BrightSource Vice President for Government Affairs and Communications Joe Desmond. He added, many are from California’s Inland Empire, where unemployment reached 15 percent at the height of the Great Recession. “These are family-wage jobs. They’re electricians, pipefitters, welders, heavy equipment operators, engineers and biologists.” There is also a helmets-to-hardhats program designed for veterans.

“People talk about the CSP industry being dormant or paused,” explained BrightSource Communications Director Keely Wachs. “Last year, PV had a banner year. The industry installed 868 megawatts in the U.S.,” said Wachs. “Now, there are 1,200 megawatts of CSP under construction in the U.S. alone, which will mean a 120 percent increase by 2013 over the 530 megawatts of operational CSP in the U.S. today.” And that, he added, “is not including the 3,000-plus megawatts under development.”

Competitors SolarReserve (which has projects under way in Nevada and California) and Abengoa (which is developing in Arizona) would agree. But BrightSource is the biggest. A 29-megawatt plant is on-line, doing enhanced oil recovery (EOR) for Chevron. The three-tower, 370-megawatt Ivanpah project, in Ivanpah, CA, near Las Vegas, is on schedule to go on-line in 2013. And both its three-tower, 500-megawatt Hidden Hills project and its three-tower, 750-megawatt Rio Mesa project are under permitting review by the California Energy Commission, with decisions expected by the middle of 2013.

Altogether, BrightSource expects to have thirteen plants, totaling 2,377 megawatts of capacity, on-line by 2017.

The BrightSource solar power tower technology was developed by the builders of the original CSP trough technology at the nine Solar Energy Generating Stations (SEGS) still in operation today -- and not far from Ivanpah.

“We produce high-temperature, high-pressure steam to turn a turbine,” Desmond explained. And, where it is contracted for, “we transfer heat from the solar field through a heat exchanger to molten salt for storage.” The technology is sophisticated enough that multinational engineering giant Bechtel was brought on at Ivanpah for engineering, procurement and construction (EPC) duties.

The Ivanpah facility without storage, said Desmond, has a 32 percent capacity factor. “A PV panel might have a 21 percent capacity factor,” Desmond went on. “If we add between two and six hours of storage, the capacity factor will be above 50 percent.” More importantly, Desmond added, “when you increase the capacity factor, you’re not necessarily increasing your cost at the same rate. You’re taking your fixed costs and spreading them out over more hours, which helps drive down cost.”

BrightSource is concerned about costs. To minimize transport expense, it has built a heliostat assembly plant at the Ivanpah facility where a union workforce turns flat mirrors into heliostats at the rate of 500 per day.

Moving to air cooling also promises cost savings. With trough technology, he explained, “there was a preference not to go to dry cooling because there was an efficiency loss.” But the tower operates “at a higher temperature and pressure that allows us to offset some of that efficiency change.”

“That is the conversion of thermal energy to electricity -- photon to electron -- efficiency,” Wachs added. “The troughs were about 36 percent. Ivanpah is 42 percent with air cooling. Hidden Hills goes up to almost 44 percent. And when you get to supercritical levels, you get to 46 percent. Like a super-efficient coal plant. That is where we are headed, because of our heliostat design and our ability to understand the sun.”

Water use is also minimized, Wachs added, because the entire system is a closed loop. The steam is condensed to water and recirculated. “We don’t require external water for cooling purposes.”

BrightSource engineers are using on-the-ground experience to find new design efficiencies. “As we go from one project to the next,” Desmond said, “we are figuring out how to do it faster, better, and cheaper. We see that already at Ivanpah. We are ahead of schedule on unit three based on what we’ve learned with work done so far on units one and two.”

A better understanding of the plant’s power block has produced a new design for the Hidden Hills project that, Desmond said, is expected to cost 40 percent more but double the output. Overall, he added, Hidden Hills is projected to be 20 percent less expensive to build.

“This is our roadmap for driving down costs,” Desmond said. “When the original SEGS plants were built here in California 25 years ago, from the construction of the first to the ninth plant, the costs came down 50 percent. Our long-term goal is a 50 percent cost reduction from where we are now. That is different from PV. They have already had an opportunity to achieve the volumes that have led to their cost reduction.”

"Until we find a technology that is low-cost, highly scalable, and long-lasting, ubiquitous grid storage won't be possible.  The all-liquid battery's elegant materials design and simple assembly process makes it the best chemical option we've seen for storing the grid at massive scale."

That's Khosla Ventures partner Andrew Chung's comment on Liquid Metal Battery Corporation (LMBC). He's now on the board of the firm; he funded founder Don Sadoway's research at MIT before the firm won the top ARPA-E award back in 2009.

LMBC just announced that it raised an additional $15 million in funding in its Round B from Khosla Ventures, Bill Gates and energy company Total. 

We spoke with Phil Giudice, the CEO of LMBC, as well. He said, "Our Liquid Metal Battery technology is tremendously exciting because it has the potential to dramatically change the electric power system everywhere," in a release. The CEO told GTM that the company had passed the R&D stage and was moving into commercializing the technology for large-scale grid applications.

The inventor of the core technology for the battery is Don Sadoway, MIT Professor of Materials Chemistry, one of MIT's most popular professors and sought-after speakers. Sadoway has challenged the research community to invent a colossal yet cheap battery. He directed researchers to look at the economy of scale of modern electrometallurgy and the aluminum smelter, which handles the holy grail of batteries: achieving a high current while maintaining massive scale.

Why is an aluminum cell not a battery? You have to produce liquid metals at both electrodes.

Making metal at the cathode is trivial, but making metal at the anode is not so trivial. So Sadoway went back to the periodic table and used magnesium to "intimidate" antimony into behaving like a non-metal. From there, using seed money from within MIT, Sadoway and his team invented the liquid metal battery or, more academically, a process called Reversible Ambipolar Electrolysis.

The battery uses molten antimony and molten magnesium separated by an electrolyte. Sadoway claims that the all-liquid configuration is self-assembling and is expected to be scalable at low cost. Furthermore, this technology may have a shot at being cheaper than sodium sulfur (NaS) batteries.

Sadoway has spoken about how rechargeables have improved as we progressed from lead acid at 35 Wh/kg to Li-ion at 150 Wh/kg (versus gasoline at 12,000 Wh/kg). But Sadoway doesn't think that Li-ion batteries have a future in grid-scale or transportation applications. We need to change chemistries and that takes radical innovation, according to Sadoway, in order to make solar and wind power more dispatchable. In this case, we need to make a battery that can handle high current.

Lithium-ion batteries in phones and cars have to be ultra-safe. Cell phones "need to be idiot-proof, largely because they are in the hands of idiots," and batteries in cars need to be able to withstand a crash. Stationary batteries for bulk storage are not held to those same requirements, which allows more freedom in choice of chemistry, but the application requires a very low price point -- and Sadoway insists that you have to think about price point at the beginning of the product design process.

It's no secret that Vinod Khosla is not a big fan of lithium-ion batteries.

Automotive traction for an all-electric car has a target cost of $100 to $200 per kWh, and stationary storage needs to be in the vicinity of $50 per kWh, according to Sadoway.

Sadoway has weighed in on the woeful state of energy research in the U.S., saying, "We need to accelerate the rate of discovery. We can make batteries two or three times better if we're willing to make the investment." He said that energy research has fallen by a factor of six, while medical research has grown by a factor of four since the 1970s. He also observes that the U.S. energy industry spends 0.25 percent of revenues on R&D, while the pharmaceuticals industry spends 18 percent and semiconductor firms spend 16 percent. Even the automotive industry spends 3 percent of its revenues on R&D.

Sadoway recommends that researchers "confine their search to earth-abundant elements. The only way to make something dirt cheap is to make it out of dirt -- American dirt."

Though revolutionary technology is a good thing, getting storage on the grid is going to involve leveraging technology, price, and, just as importantly, regulatory issues involving the FERC, and ISOs and PUCs across the nation. 


 

The usual way to fix that is to put in another base station or concentrator, at additional cost. That’s because routing the network around changes in its environment is complicated, both to plan and then to deploy -- and once you’ve deployed, the trees have gone and grown and fallen again, wrecking your model.

That’s how Andres Carvallo, chief strategy officer at Proximetry, described the problem he’s hoping a new partnership with EDX Wireless will solve.

In simple terms, Eugene, Ore.-based EDX makes the tools to plan networks, and Proximetry makes the tools to manage them, Carvallo said in a Tuesday interview. The San Diego, Calif.-based startup’s software runs on its own, or via licensing partners like Cisco, which uses it for its smart grid network management offering, or CSC Corp. for a cloud-based network management service for utilities.

From there, Proximetry feeds all that data back into the EDX planning process, where it goes into a new cycle of modeling, he said. That keeps engineering and operations updated with fresh tools to make plans, fix problems, and catch the slow degradation of performance that would otherwise manifest in failure and expensive upgrades.

Carvallo saw the problems that neighborhood networks can experience firsthand during his stint as CIO of municipal utility Austin Energy, where crews trim some 400 linear miles of power line corridors per year, mostly to prevent branches from touching power lines and starting fires.

But it turns out that leafy green residential neighborhoods make for poor wireless connectivity as well. At Austin Energy, Carvallo estimated that up to 15 percent of network devices -- smart meters -- experienced failure at one point or another due to such changes as vegetation growth, new construction, public works projects and other environmental changes.

“Today, it’s all manual troubleshooting” to fix the problem, he said. “You have to send people there to figure it out.” Proximetry can optimize networks in a way that avoids having to reinforce troublesome networks with personal inspections and new gear, to the tune of 30 percent to 40 percent reductions in those network maintenance costs, he said. The software can also detect imminent failure for preventative maintenance, or notice when gear is running just fine and doesn't need to be replaced on a schedule, all for additional savings and operations benefits.

Proximetry isn’t alone in offering smart grid network management of some kind, of course. Telcordia, now owned by Ericsson, has a smart grid NMS offering. Another contender is GridMaven, a business unit of SK Telecom’s American subsidiary that launched in January, the same day as Cisco announced its Proximetry-backed network management system. This week, GridMaven announced it had finished deploying the network management system for South Korea’s Jeju Island, a national smart grid test bed that’s one of the biggest single deployments in the world.

But for now, GridMaven has said it is sticking to network fault detection and correction, rather than automation of network responses. Proximetry, on the other hand, is looking at increased automation with EDX, in terms of constantly updating the models that utilities use to plan and execute their projects.

“There’s got to be a closed loop, iterative cycle of design, planning operations, optimization, design, and so on and so on,” Carvallo said. Otherwise, every new project needs a team of engineers to solve all the new problems, he said.

Just how Proximetry’s technology compares to the likely contestants in the field of winning utility business remains to be seen. The big cellular carriers have their own network optimization underway to support the smart grid’s particular needs, for example. Smart grid services offerings from the likes of General Electric, SAIC and Lockheed Martin presumably have some way to tie their views together.

Proximetry was founded in 2005 and has been backed by Munich Venture Partners, Aeris Capital, Investec, and Rembrandt Venture Partners. It raised a $5 million Round A in 2007 and in June raised $792,000 of a planned $1.08 million round, according to a regulatory filing. It has also won grants from European government entities (PDF), and has offices in Poland and Germany.

The association reports, for instance, that in 2010 geothermal energy generated “twice the amount of electricity as solar energy did worldwide.” The world’s installed grid-connected PV capacity went from 7.4 gigawatts at the end of 2009, to 16.8 gigawatts at the end of 2010, to 29.7 gigawatts of grid-connected PV at the end of 2011.

The world’s installed geothermal capacity, as of May 2012, was 11,224 megawatts (11.2 gigawatts), up from 10.7 gigawatts of installed capacity in 2009 (according to the IEA). It would appear geothermal did have the lead sometime in 2010, but comparing the sectors' growth rates is not flattering to geothermal. (Geothermal obviously has an advantage when it comes to generating kilowatt-hours due to its higher capacity factor.)

The telling measure is geothermal’s U.S. performance. According to the GEA report, utility-scale geothermal originated in the U.S. in the 1960s. The U.S. “remains the world leader with approximately 3,187 megawatts of installed capacity,” but the GEA reported it brought only “approximately 91 megawatts of capacity on-line between 2011 and early 2012.”

That’s far less than the 1,855 megawatts of photovoltaics deployed in the U.S. in 2011.

The U.S. wind industry added 6,816 megawatts (6.8 gigawatts) of new capacity in 2011, a 30 percent increase over the previous year’s growth.

There is, however, cause for optimism in the geothermal industry. While GEA President Karl Gawell noted that “growth in the United States is still hindered by uncertainty about the direction of government policy” (an observation echoed by leaders in the solar, wind and natural gas sectors in 2012), the rest of the world plans to seize the geothermal opportunity.

The Turkey Geothermal Association estimates, according to GEA, that its industry will grow from 100 megawatts of installed capacity to 500 megawatts by 2015. Kenya, with 202 megawatts of installed capacity, is developing fourteen new sites. And Indonesia, which is situated on the famed ring of fire and is estimated to have 27,510 megawatts of geothermal potential, intends to build its installed capacity to 5,000 megawatts by 2025.

A recent GTM interview with EnergySource President and CEO Dave Watson, who just commissioned the 49-megawatt Hudson Ranch I project in California’s potential-rich Salton Sea known geothermal resource area (KGRA) and is already readying a matching 49-megawatt Hudson Ranch II project, offered insight into the GEA’s positive attitude.

“At some point,” Watson said of grid operators now buying up solar and wind contracts, “they’re going to want some grid stability [and] we’re running twenty-four hours a day. When the utilities get tired of dealing with intermittency and want baseload renewables, there’s really only one source -- and that’s geothermal."