There is some real history of solar power for large scale power generation – and it isn’t on a rooftop or out in space. The first major solar thermal power plants went into the Mojave Desert in 1984 and have been producing over 330 MW since. There has been a major upswing in installations of concentrating solar power (CSP) recently, and an even bigger increase in the announcements of projects – 8 GW is in the pipeline.
However, there are some serious issues which may spoil the party for CSP. The immediate challenge is to avoid another bust by turning the GWs announced into actual GWh of electricity, proving its bankability and capability.
The CSP industry began in the late 1980s and early ‘90s in California with the construction of a series of parabolic trough plants in the Mojave Desert. The first CSP plant came online in 1985 with Solar Energy Generating System #1 (SEGS I) with a capacity of 13.8 MW.
Subsequently, there were 8 other SEGS plants which were built in the same region of the Mojave Desert, with the last one coming online in 1991. The total output of the collective set of parabolic trough plants is 354 MW. And, according to Hank Price, one of the original engineers working on the project, these early projects offered the chance for not just technology evaluation, but also the opportunity to study costs and possibilities for cost reduction.
Indeed the cost of wholesale solar thermal electricity from the SEGS plants dropped from US$0.24/kWh to under half of that. The original project has been generating electricity for three decades and has provided 14 TWh of energy.
Talking to Price and his colleague Tandy McMannes, who is another of the original team members, one can feel their sense of pride of being part of homesteading the Wild West and setting up the first CSP ranch. McMannes explains how it wasn’t until after SEGS III, that they were finally able to attract investors including multiple eastern utilities. He helped negotiate the Standard Offer Contracts, which fixed power purchase rates and which had just been introduced in California in 1983 for wind energy. In addition to setting up the original solar farm producing power at contract rates, they demonstrated capability and bankability (although it wasn’t called that back then).
Boom and Bust Cycles
Although successful by most metrics, the original company that developed the project – Luz – had to declare bankruptcy. There were multiple forces at work which discouraged solar power.
As oil prices fell and remained low through the 1990’s, construction of new solar plants stalled. The regressive energy policy starting with the Reagan Administration, along with the vagaries of property taxes, forced the company to close down but the investor groups continued to sell electricity to utilities.
Even with the early successes, CSP was basically abandoned in the 1990s and it wasn’t until the mid-2000’s that new companies began looking again at innovative ways to generate large scale solar power.
The oil prices rose once more and there was an increased focus on reducing carbon emissions, as well as new renewable portfolio standard (RPS) levels in various states. As standard solar photovoltaic (PV) panels were still expensive and therefore not suitable for utility scale power generation, there was a renewed interest in solar thermal as a viable option.
A second cycle of activity began in 2004 with a revival of greentech in the Silicon Valley. Luz Founder, Arnold Goldman, regrouped his old team, and founded BrightSource Energy. In addition, Silicon Valley venture capitalists gave some substantial funding to new-comers such as Ausra and encouraged the renaissance in CSP.
Details of the SEGS Ranches
The SEGS plants range in capacity from 13.8 MW to 80 MW, and they were constructed to meet Southern California Edison Company’s periods of peak power demand.
The plants operate for 80% of the summer midpeak hours and 66% of the winter mid-peak hours. A natural gas backup system supplements the solar capacity and contributes up to 25% of the plants’ annual output.
The SEGS plants use parabolic trough solar collectors to capture the sun’s energy and convert it to heat. In the SEGS design, the curved solar collectors focus sunlight onto a receiver pipe. Mechanical controls slowly rotate the collectors during the day. The collectors concentrate sunlight 30 to 60 times the normal intensity on the receiver, heating the heat transfer medium – the synthetic oil – as high as 735°F (390°C).
The heated oil is routed through a heat exchanger to generate steam that drives an electricity producing turbine.
Learning from the early SIGS plants has been extensive. During the 1990’s the SEGS’s technical team worked with the Department of Energy’s (DoE) Sandia National Lab to develop cost-reduction strategies for O&M planning optimization, subsystem automation, mechanical and reliability improvements and overall performance improvements. Many of the cost reduction strategies can now be applied to other CSP technologies and the industry in general. The combined efforts developed capabilities for reducing O&M costs by a third. Note that O&M is a substantial part of CSP running costs unlike fixed panel solar photovoltaics (PV), where the O&M costs are a small fraction.
In addition, the plants were able to perform to exceptional standards and achieved a record output for single-day generation of over 2 GWh, and established a world record solar thermal to electric efficiency of 18%. The SEGS 1-9 configuration is a true success story in the annals of CSP.
Another set of companies began to push back into the market and re-engineer the CSP solutions for a new cycle.
What is Happening Now?
There are now almost 8 GW of CSP projects in the pipeline, compared with the 500 MW that will be installed by the end of 2010. This project pipeline is in stark contrast with recent news about cancellations of CSP power plants.
What is going on? Basically, the CSP industry is at a crucial turning point due to a number of critical factors. The ‘potential’ of CSP is an oft-heard term, and it is true that there are some very attractive aspects of CSP – in principle. CSP has two key positive aspects: It can provide good energy dispatchability, and the ability for hybridization.
The two elements are interrelated. CSP can be configured to have thermal storage and can provide up to 6 or even 8 hours of post-sundown energy. Hybridization allows the power plant to have both solar and standard energy for heating. In principle, this capability would allow for 100% capacity – where the power plant is providing electricity over the full 24 hours, and the plant becomes a tunable baseload supply – the Holy Grail for renewables.
Storing CSP Power
Indeed, there is a significant amount of research going into CSP storage and multiple worldwide efforts are underway to find a low-cost and flexible solution. Unfortunately, for the near future, the utilities and the market don’t provide a value for this storage and it is still in the R&D phase, and consequently, still expensive.
The ability of hybridization is important because much the notion of a fully electric car, the hybrid car, is an important step to that ultimate goal.
Hybridization is important for solar in that it improves the intermittency factor, the shaping of the electricity, and the capacity factor. Again there are some limitations, however.
Ironically, if solar is paired with a carbon-based power source, it can have two negative effects: Solar projects are now limited to 2% natural gas in order to remain designated as ‘renewable’ and hence qualify for support. (The SEGS I-IX projects are limited to 25%). Second, in some instances, the use of a coal-fired plant, if run at less than full capacity during times of high renewable generation, it can become substantially less efficient and actually contribute to higher air pollution.
Clearly, these issues are much intertwined.
The good news for now is that the US Department of Energy (DoE) is ramping up its CSP research, development, and deployment efforts, leveraging both industry partners and the national laboratories. DoE’s goals include increasing the use of CSP in the United States, making CSP competitive in the intermediate power market by 2015, and developing advanced technologies that will reduce system and storage costs, enabling CSP to be competitive in the baseload power by 2020. In addition, there are numerous support programs at the National Renewable Energy Laboratory (NREL) for storage projects.
Luz/BrightSource Shines Again
More good news is that BrightSource has broken ground on another project in the desert at Ivanpah, 40 miles south west of Las Vegas, NE. The facility, scheduled to come online in two years, will have a capacity of almost 400 MW, making it the world’s largest CSP plant currently under construction. The project is a combined effort between BrightSource and none other than Bechtel (the largest engineering company in the USA, which built the Hoover Dam in 1933). A key element to this project’s success was securing a US$1 billion Loan Guarantee from DoE.
One of the encouraging aspects to this project is that it was able to balance various competing requirements from both an engineering, environmental impact, and advanced design perspective. Because of legitimate concerns about environmental disruption, the engineers designed a system which minimized impact by almost eliminating the need for extensive land grading and concrete foot pads. By placing individual mirrors on poles directly into the ground, the system allows vegetation to co-exist within the project and avoids impacting sensitive habitats. Furthermore, to conserve scarce desert water, the tower uses air-cooling, which compared with conventional water-cooling, results in 90% less water usage, although with some loss in overall efficiency. The water that is used is part of a closed loop system within the boiler segment.
In this regards, the Ivanpah project is a good template for CSP developments as it includes multiple stakeholders such as the developers, utilities, and environmental groups. This mix is often reflected at CSP conferences, where that very mix of individuals is asked to contribute to the discussion recognizing that large-scale projects must be carried out in an environmentally responsible manner.
However, there is some bad news as well. Even in the context of multiple 100s of MWs of CSP in the pipeline, there have been some notable cancellations. One recent example is that of a 92 MW CSP facility that El Paso Electric Co had planned for Santa Teresa, NM. The project collapsed because of difficulties with securing financing. The utility now plans to build a 20 MW PV facility instead.
Issues: Cost and Bankability
Despite the upsurge in the CSP pipeline, there are some warning signs that CSP still has the potential to hit another bust cycle. The issues are an unfortunate perfect storm of three elements: The primary one is the ‘bad luck’ of the global economic squeeze which has made the access to capital very restricted.
CSP and Storage: A Moving Target
CSP can potentially provide the needed dispatchability with storage; however, the rest of the renewable industry is not standing still. Recently, the town of Presidio, Texas, brought online a huge battery bank to allow for wind power storage and provide the town of 4000 inhabitants with up to 8 hours of uninterrupted power. The initial costs are high – US$6/W – but this is just the start of the learning curve.
According to SolarVision’s projections, if this cost drops along an accelerated learning curve for five years; and when PV installations are paired with storage, the full cost of dispatchable PV power becomes US$4/W, which is nipping at the heels of CSP with storage. Note that there is a lot of analysis that gets folded into these simple metrics, such as O&M costs, LCOE comparisons, and dynamic disruptions to the learning curve, so the numbers are meant to be illustrative.
The second element is the increasing drop in the prices of solar PV and the ever expanding track record for large-scale PV installations. The third element is that CSP projects are typically large-scale and require major amounts of not just money, but time for building, and regulatory review.
For the first point, there is a very significant question about bankability which has become a common buzzword on the solar arena. This is a legitimate concern as investors don’t want to take on risk for less-than-fully proven technologies.
When you couple that uncertainty with the large investments needed for CSP, the result is major hesitation. For the case of El Paso Electric, the story is a perfect example of a developing storm on the open desert planes.
David Knox of NRG Energy Inc, which El Paso Electric contracted to develop the CSP plant, explains the situation in simple terms. NRG sought a Loan Guarantee from DoE to build the facility, but its request was rejected. Rather than wait for NRG to get funding, El Paso Electric switched to solar PV to meet the state’s renewable energy mandates.
“We can get banks to finance a solar PV project now, but they’re not yet ready to finance a solar thermal project,” Knox says. “That’s why we need to rely on Loan Guarantees to do it, but we won’t get that in the time frame that El Paso Electric needs it.” Jason Marks of the New Mexico Public Utilities Commission (PUC), adds: “The problem is not just timing – it’s an endemic situation facing the CSP industry”.
It is becoming apparent that a prerequisite for any CSP project is a DoE Loan Guarantee; a problem considering the growing dissatisfaction with further US Government support in the form of funding by the newly elected Republican House majority in the House of Representatives, and the emotional wave of budget contraction policies promoted by a segment of the American populace.
In addition, there have been other recent cancellations of projects where the utilities switched from CSP to solar PV citing costs. The dynamics here are evident as seen in Figure 1 where the learning curve of PV is much steeper and advanced than that of CSP, let alone CSP with storage. This situation becomes a reinforcing cycle: Cheaper PV leads to more installations, which in turn leads to still cheaper PV. This is the essence of the solar market, getting the cost down and moving along the learning curve as quickly as possible.
Another problem is that according to SolarVision’s analysis, the often touted metric of Levelized Cost of Electricity (LCOE), which is often used to promote CSP, is becoming lower for PV and is projected to continue on this trajectory.
Is Bigger Better – Or is Small Beautiful?
So where are the areas of opportunity for CSP to avert another bust? There are some developing trends which could provide a path to viability: Getting bigger by going smaller.
There are emerging market opportunities in developing countries with a poor or nonexistent grid and power infrastructure. In these countries, such as India, the need for local power is great, but the regional demand is only on the scale of a few MWs.
This application is a relatively hidden market but it has been estimated to be possibly on par with the nominal ‘grid-connected’ market. In some regards, this market is similar to the cell phone model where the landline infrastructure is not in place, but local service can be provided by towers. The demand here is great, and the need for viable energy which provides both immediate as well as some storage is necessary.
The ability of CSP to provide both hybrid as well as storage gives it an edge for the time being. However, there is a big caveat, which is the necessity for micro-CSP. Typical CSP installations need to be large-scale to amortize the balance of plant (BOP), and the O&M costs, so minimum sizes are in the multiple 10s of MWs, much too large for micro-grid requirements.
However, two companies are examples of the thinking to address this requirement: One is Aora, and the other is Sopogy. They are both striving to provide sub-MW CSP solutions. Sopogy is based in the USA, has been around since 2002, and it already has a 2 MW project operating in Hawaii with new microgrid contracts underway. Aora is based in Israel, and is targeting to get a 100 kW demonstration unit in Spain by mid-2011.
A problem with large sizes is that although there is the argument of economies of scale, there are also other costs and barriers which increase with size. Permitting is more complex, labor rates can go up, and the need for grid upgrades or transmission can be substantial.
One answer to averting a bust-cycle is simple: Deliver on a major number of the projects currently in the pipeline. It’s time to put-up or shut-up and demonstrate CSP’s potential.
There are some leaders in the CSP segment who recognize the urgency and are pushing for re-focusing energy into execution and co-operation. For instance, Stephen Mullinnex of US Renewables Group which is backing the CSP installer SolarReserve will say that his competition (e.g. BrightSource) is not really the competition – because the more CSP installed, the better for everyone in the industry.
As he puts it: “It’s now more about MWh than about ‘bragga-watts’.” In other words, we need to act swiftly to execute projects in the pipeline rather than bringing more volume to the pipeline, because time may well catch up with you.
CSP has gone through some boom and bust cycles in its 30-year history. It is poised to make a comeback but there are some significant clouds gathering on the horizon. For CSP there is the urgent need to prove its future capability (i.e. its potential) now.
The CSP providers need to achieve significant momentum. In an unfortunate storm of market conditions, CSP is getting squeezed between tighter capital and the plummeting costs of PV. It would be ironic if the iconic origins of terrestrial solar power, the SEGS plants, remain just an image of a CSP boom-bust cycle in the Mojave Desert, still cranking away on a tumble weed-blown plane, but left behind because of market forces and the dynamics of technology. Not the scene from the Western movie with no sequel that we want to have playing.
About the Author:
Andy Skumanich is the Founder and CEO of SolarVision Co.
Renewable Energy Focus U.S., Issue 3 November/Demcember 2010.