A new solar concentrator design developed by an electrical engineering PhD student at the University of California, San Diego, could lead to the development of inexpensive solar concentrators that require fewer photovoltaic cells than existing designs.
Existing solar concentrator systems typically use arrays of individual lenses that focus directly onto independent photovoltaic cells, which all need to be aligned and electrically connected.
In contrast, Jason Karp’s new-design solar concentrator collects sunlight with thousands of small lenses imprinted on a common sheet. All these lenses then couple into a flat waveguide, which funnels light to a single photovoltaic cell.
Karp built a working prototype with just two primary optical components, therefore reducing the cost of the optics associated with the entire system. What is more, his solar concentrator is compatible with high-volume, low-cost manufacturing techniques such as the existing roll-to-roll processing techniques used to make large televisions.
Karp designed and built prototypes of the new solar concentrator in the Photonic Systems Integration Laboratory led by electrical engineering professor Joseph Ford from the UC San Diego Jacobs School of Engineering.
He is currently examining variations of the basic structure of the concentrator to increase both concentration and optical efficiency.

Individual secondary optics require multiple PV cells (a). A slab waveguide homogenises and transports sunlight from all apertures to a single cell (b). Increasing the waveguide length does not increase the required PV cell area. Arrows indicate PV cell locations.
A very brilliant idea. I hope, that the implementation can be realized shortly. I do look forward to hearing the news that this “CPV” will be available for sale.
Wish you a successful progress.
Reading the information from the site (http://psilab.ucsd.edu/research/slab_concentration/main.html ) it looks like it only has 80% optical efficiency at 240x. It also didn’t explain what happens with different wavelengths or describe the acceptance angle for the light. Concentrating is definitely the way to go since you use much less material and can get much higher efficiencies. I think a simpler more efficient approach like http://www.sol-solution.net/Technology.html shows more promise.