Google opensourced their Heliostat Optical Simulation(HOPS) software, which can be used to simulate a field of heliostats and a receiver that provides thermal input to a heat engine. The code is available at the following location.

http://code.google.com/p/hops/

Power towers are a great way to convert solar energy into electricity. Google intended to use Brayton Engines with the Heliostats they developed, but Google eventually stopped funding the project. The problem with Brayton engines is that they typically are not efficient until a very high input temperature is reached, and they typically are not as efficient as Stirling engines. The very high temperature required by Brayton engines became problematic for receiver design. The receiver needed to withstand average temperatures of 1000 Celsius. Google claimed to stop working on solar thermal applications due to the problems they encountered with receiver design and the falling cost of PV cells. In contrast to Brayton Engines, Stirling Engines do not experience a massive drop in efficiency as temperatures drop.

It seems that large Stirling engines combined with a sodium reflux receiver is a cost effective method that could compete with PV cells and other forms of solar thermal electric production. Stirling engines have been built that are 50% efficient at converting thermal energy to mechanical energy when operating between temperature reservoirs of 750 Celsius and 50 Celsius. I have design plans for one such engine that has 10kW of electric generation capacity. A physical,
working prototype has been built, and I believe this engine can be scaled in size to 300kW and would weigh around 2700kg. Sodium reflux pool boilers have been developed by Sandia National Laboratories and simulation software
has been designed(CIRCE2/DEKGEN). However, the SNL code is closed source and only available to Federal employees and Federal contractors. The method, however, is discussed in books/journals. These reflux receivers have been tested for 7500 hours. I would like to design a scaled up version of one of these receivers. Perhaps something like 3.5 MW
thermal. I believe such a receiver would be a little over 94% efficient. The 75kW version was 92.8% efficient. The increased efficiency comes from lower convection losses. One of these receivers could be connected to a few 300kW or larger Stirling engines.

Another company that uses Power towers is Bright Source Energy. The problem with Bright Source Energy's technology is that they utilize steam. Steam has to be heated in a boiler whose material requirements scale as a linear function of pressure and of volume. Due to the fact that steam's thermal conductivity is between 3.3 times(at 1200K)  and 8.28 times(at 300K) lower than that of helium, it must stay in the boiler for a longer period of time than a gas such as helium in order to soak up the heat. As a result, Bright Source Energy will have greater material requirements per kilowatt of generating capacity than an approach based on large Stirling engines and sodium reflux receivers.

Any comments or interest in the project?