From: rrllff@yahoo.com   
      
   On Monday, January 13, 2014 12:02:06 PM UTC-5, haitic...@gmail.com wrote:   
   > As a thought experiment, consider that an average Si PD detector is somewhere   
   around 2mmx2mm, as I remember. So the question arises in my mind, has anyone   
   cut   
   the rather expensive filters into sizes needed for the detectors? I looked   
   around, and Tungsten carbide scribe goes for $8 (seems to work better than the   
   diamond variety.)   
   >   
   > This could be done by hand, but a "scribing engine" worth consideration. I'm   
   still thinking about it, but ideal is to have tensioner spring, guide for   
   sliding, and way to immobilize the filter.   
   >   
   > What is needed for immobilization is a glue to temporarily fix the filter.   
   The   
   dopping wax used by diamond polishers, the old standby Duco cement, or Sodium   
   Silicate, a water soluble glue. You probably dont want to melt wax, or expose   
   filter to water, so that leaves Duco released with acetone. I'm even nervous   
   with acetone, since it has a hydrophilic double nature capable of penetrating   
   inorganic films. I'd feel better with a strictly hydrophobic solvent like   
   toluene, methylene chloride, or hexane.   
   >   
   > Possibly someone will post that the filter manufacturers like OCLI are doing   
   this already, but given that detectors are generally small area, it's a bit   
   surprising to me that the filters are so large.   
   >   
   > JB   
      
   If you wanted a large number of small filters it probably makes more sense to   
   load up the coating machine with parts the size you want.  They can then be   
   edged sealed individually.  (Note that tooling will not be cheap.)   
      
   When you cut down a larger filter, the cutting process will result in edge   
   damage, extending deeply into the clear aperture (for 2mm x 2mm).   
      
   In many cases it is simpler and cheaper to just use a larger filter.   
      
   Also note, that if the filter is used in image space (near the detector) it   
   will   
   be in a space with high angles of incidence.  Depending on your application,   
   this may make it impossible to use this configuration for your application.   
      
   http://www.richardfisher.com   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)