Its been a while since I said I would have some results real soon. I've been occupied with another project of mine (which, by the way, I am *very* excited about), but decided to bite the bullet and put this up.
It is not as exciting as the funnel paper but It demonstrates concretely ideas we have been discussing. Without knowing the details of how large curved reflectors would affect the cost of a system compared to small flat reflectors, the models demonstrated here show the large curved reflectors to be superior per system foot print. Also, the utility of the compound array is clear.
Future work would explore the sensitivity of a design to inaccurate tracking. And a compound array should be explored. Its only about a days worth of work to accomplish. Mostly, a lot of waiting for 1500 rays to be calculated for each of many different angles and then realizing I made the entry aperture too small or something and having to run it again.
For now though, I am going back to my other project.
This tool is available for use when the turbine project, or other sympathetic projects, are ready to take more steps.
waiting is fullness,