During three years of remarkable progress at Ausra and AREVA Solar, I was fortunate to lead an incredible team and to have the backing of a CEO who trusted my ability to deliver innovation into the field. Those years were immensely gratifying, and it was the best job I’ve ever had!
Four years after the French nuclear giant AREVA S.A. acquired the venture backed startup Ausra, Inc., it seems an opportune time to reflect on the evolution of CLFR technology from low pressure saturated steam to high pressure superheated steam.
The early CLFR units constructed by Ausra’s predecessor, Solar Heat and Power PTY Ltd at the Liddell Power Station in New South Wales, Australia were considered to be v1 and v1.1. Those and the three v1.2 units constructed by Ausra in 2008 at the Kimberlina Power Station in Bakersfield, CA were all saturated steam systems, at low to moderate pressures. As Director of Power Engineering, I was involved in the construction, and commissioning of the Balance of Plant systems at Kimberlina in 2008, but I was skeptical that the saturated steam system envisioned by Ausra would be economical, or even practical from an operations perspective.
I advocated an entirely different approach — Once Through to Superheat — which would eliminate a substantial amount of Balance of Plant equipment to reduce cost, increase efficiency, and simplify operations. In February 2009, CEO Bob Fishman put me in charge of Engineering and gave me a free hand to choose my team to make this happen.
At the time there was widespread belief in the solar industry that CLFR technology could not deliver superheated steam because its concentration ratio (mirror to receiver area of 30), was insufficient to overcome the radiation losses at superheated steam temperatures. It was felt that a high temperature selective coating with low emissivity and high absorbtivity was needed, but there were no air-stable coatings suitable for the boiler tubes used in the Ausra system.
However, Kimberlina testing showed the flux on the boiler tubes was highly non-uniform and peaked, and although the average concentration ratio was 30, the peak concentration was sufficient for superheated steam. Accordingly, tubes near the peak flux were selected to produce superheated steam, and the others used for feedwater heating and evaporation. Throughout the 2009 operating season at Kimberlina, my engineering, construction and operations teams modified the three units commissioned in 2008. By October, they had achieved what had previously been thought impossible: sustained superheated steam, even through intermittent clouds! This is shown below in a chart I presented at the 2010 CSP Today conference.
In parallel with the modifications made to the v1.2 systems during 2009, the Engineering team also completed a v1.3 design release with technical, cost and performance improvements, and Ausra gained boiler design certification under the ASME Section I ‘S’ stamp program.
As VP Engineering and Chief Engineer, I was responsible for Ausra’s CLFR (Compact Linear Fresnel Reflector) product and technology roadmap. The CLFR roadmap showed a trajectory to deliver the lowest Levelized Cost of Energy (LCOE) of any solar thermal system. I felt that our product and technology roadmap would ultimately deliver solar thermal price-performance leadership. Some of the key elements underpinning the LCOE trajectory were:
- the potential for higher cycle efficiency than parabolic trough by producing steam at higher pressures and temperatures,
- lower Bill of Materials cost than trough or tower by use of flat glass, ganged reflectors driven by a few positioners, and supported on fewer foundations, which was a natural consequence of the compact geometry,
- lower construction cost than trough or tower due to less civil work (grading and foundations) because of land efficiency.
Nevertheless, realizing the roadmap’s potential would require more capital than the VCs would commit, particularly during the depth of the financial crisis then underway, and a strong balance sheet to offer meaningful guarantees to customers. Accordingly the merger of Ausra into AREVA was both timely and necessary.
The next milestone on the roadmap was to beat the parabolic trough competitors by matching their steam conditions (100 bar at 370C) while beating them on cost. My conviction that the v1.3 design combined with the superheated steam capability would deliver a lower LCOE than the parabolic trough competition was confirmed when Ausra won a competitive bid for a solar booster at a California utility.
But AREVA’s purchase price depended on confirming that the v1.3 design actually met the roadmap LCOE target — paper studies were not enough. The LCOE was to be verified by an independent engineer, based on Bill of Materials (BOM) cost, thermal performance, construction Cost and Schedule, transient behavior, and reliability, as demonstrated in a new build Solar Steam Generator — SSG4.
SSG4 was designed, constructed, commissioned and tested by November 2010. It was a complete success, resulting in a substantially lower LCOE than the roadmap target. The design, construction, commissioning and testing of SSG4 were described in a series of four papers presented at the 2011 ASME Power Conference.
- Superheated Steam From CLFR Solar Steam Generators
- Control of Solar Steam Generators
- Integration, Commissioning and Operation of Solar Steam Generator 4 at Kimberlina
- Thermal Performance Testing of Solar Steam Generator 4 at AREVA Solar’s Kimberlina Demonstration Facility
The successful SSG4 project also became a catalyst for commercial awards for the CLFR technology:
- The Kogan Creek Solar Booster, a 44 MWe add-on to the 750 MWe supercritical plant owned by CSEnergy in Queensland Australia;
- The 250 MWe Solar Dawn project in Queensland, which despite more than $500 million of federal and state government grants was cancelled by AREVA;
- The 2 x 125 MWe Reliance project in Rajasthan, the world’s largest solar power plant to be registered under the United Nations Clean Development Mechanism.
- The 5 MWe Sundt Solar Booster at Tucson Electric Power’s plant.
My AREVA Solar Engineering team continued to innovate at a rapid pace by combining Silicon Valley style Rapid Product Development with SkunkWorks techniques. I’ll blog about that another day. But right on schedule, on July 1, 2011, we released a v2 platform design that realized an additional 25% LCOE reduction through optimization, per the Roadmap. This was followed just six months later by the v2.1 design release, which achieved a further LCOE reduction with enhanced maintainability and operability. This was the final CLFR design for which I was responsible at AREVA Solar, and is currently being commissioned in India.