From: hwabnig@.-   
      
   On Sat, 22 Oct 2011 09:43:56 +0300, Mikko OH2HVJ    
   wrote:   
      
   >Hi,   
   >   
   >We're building a device with a light source that has multiple LEDs   
   >combined to give suitable wavelength distribution. Earlier I asked   
   >here advice for doing this.   
   >   
   >We ended up with a structure where  the LEDs are mounted on a PCB,   
   >surrounded with high-reflectivity foil. The LEDs are shining on   
   >a semisphere from the same material and the exit fiber is mounted   
   >in the middle of the LEDs. Kind of an integrating sphere, the   
   >wavelenght distribution, spatial form are excellent and the   
   >solution is very stable and easy to manufacture. Also, we've   
   >got about 90% of the theoretical coupling maximum, which is   
   >surprisingly good for a quick prototype. (The theoretical coupling   
   >maximum is nothing to write home about..)   
   >   
   >Ok, now we've got a good working prototype which fulfills the   
   >original measurement target.   
   >   
   >The customers product manager has some technical background and keeps on   
   >suggesting different solutions which might in his opinion improve the   
   >efficiency.  I'm usually very good at explaining complicated technical   
   >issues to people of different leve, but in this case I feel completely   
   >unarmed!   
   >   
   >So, the question is: how would you explain the idea of optical invariant   
   >as simply as possible ?   
      
      
   What is an "optical invariant"?   
      
      
   w.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)