On Tuesday, April 8, 2014 5:09:47 PM UTC-4, Phil Hobbs wrote:   
   > On 04/08/2014 04:30 PM, ggherold@gmail.com wrote:   
   >   
      
   >   
   > That's the point of them.  They gather up etendue at one angular   
   > acceptance and deposit it at a wider one, but in a smaller area.   
   >   
   > The design principles are actually pretty cool--for an expert case, I   
   > recently got a copy of Welford & Winston's book, which is really   
   > illuminating. ;)   
   >   
   > The starting point is the edge-ray principle, which says that in order   
   > not to spill light out of the detector area, you want the rays at the   
   > edge to be in perfect focus, but you don't care about the other ones.   
      
   Yeah that's exactly what we did.  We made cones from stainless, then   
   electroplated copper, potted in epoxy, and removed the copper light cone.   
   Cone angle and final hole size, picked as you say from tracing the edge rays.   
      
   Hey this a is wonderful.  I listen to you talk about etendue and understand   
   sorta.. but now I've got a real world example to stick in my head!   
   Great!   
   >   
   >   
   >   
   > That's how 1-D CPCs work--the right hand edge of the detector is at the   
   > focus of the left-hand parabola, and vice versa.  The light is all   
   > scrambled up in between, but in sharp focus at the edge, so none gets   
   > lost.  That's why they're 100% efficient in 1-D.   
   >   
   >   
   >   
   > You can do it in 2-D by putting plane mirrors at the ends of the trough,   
   > which is ideally still 100% efficient, or by taking the 1-D design and   
   > making a figure of rotation.  (The compound paraboloid design isn't   
   > actually a paraboloid--the axis of the rotation isn't the parabola's   
   > axis of symmetry.)   
      
   A colleague made one (a light cone) that was parabolic.  I think the reason   
   was that it saved space (length wise) when compared to the straight sided cone.   
      
   George H.   
   >   
   >   
   >   
   > There are some high-angle skew rays that rattle round the inside and go   
   > back out without hitting the detector, so it isn't exactly 100%, but a   
   > good design gets pretty close.   
   >   
   >   
   > Cheers   
   >   
   >   
   >   
   > Phil Hobbs   
   >   
   >   
   >   
   >   
   >   
   > --   
   >   
   > Dr Philip C D Hobbs   
   >   
   > Principal Consultant   
   >   
   > ElectroOptical Innovations LLC   
   >   
   > Optics, Electro-optics, Photonics, Analog Electronics   
   >   
   >   
   >   
   > 160 North State Road #203   
   >   
   > Briarcliff Manor NY 10510   
   >   
   >   
   >   
   > hobbs at electrooptical dot net   
   >   
   > http://electrooptical.net   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)