From: pcdhSpamMeSenseless@electrooptical.net   
      
   On 2/1/2014 6:12 AM, haiticare2011@gmail.com wrote:   
   > On Monday, January 27, 2014 10:35:29 AM UTC-5, Phil Hobbs wrote:   
   >> On 01/27/2014 02:20 AM, Mikko Syrjalahti wrote:   
   >>   
   >>> Phil Hobbs  writes:   
   >>   
   >>>>>>>>>>>> Hi, all,   
   >>   
   >>>>>>>>>>>>   
   >>   
   >>>>>>>>>>>> I have a gig to design a microplate reader for   
   >>>>>>>>>>>> a new bioassay system.   
   >>   
   >>>>>>>>>>>> To match the reagent systems, it needs to work   
   >>>>>>>>>>>> over a range of   
   >>   
   >>>>>>>>>>>> wavelengths in the 340-500 nm region, none of   
   >>>>>>>>>>>> which is particularly   
   >>   
   >>>>>>>>>>>> well   
   >>   
   >>>>>>>>>>>> matched to mercury emission lines.   
   >>   
   >>> ...   
   >>   
   >>>> Target is +- 0.0002 absorption units over a 2 AU range.  Not   
   >>>> difficult   
   >>   
   >>>> from a SNR point of view, but you don't get stability like that   
   >>>> by   
   >>   
   >>>> accident.   
   >>   
   >>>   
   >>   
   >>> We've done a project that had specs close to this; 0-2 OD at   
   >>> 0.0001 and   
   >>   
   >>> going to 3.5/0.05 400-700nm. We ended up combining multiple LEDs   
   >>> with a 7:1 fiber   
   >>   
   >>> bundle. Loses a lot of light, but the measurement package needs a   
   >>> fiber   
   >>   
   >>> coupled source anyway. If I remember correctly, we had some dips   
   >>> at 430   
   >>   
   >>> and 480nm due to not finding suitable high power LEDs.   
   >>   
   >>>   
   >>   
   >>   
   >>   
   >> I may well wind up with something a lot like that.  (Great minds   
   >> think   
   >>   
   >> alike, and fools seldom differ, as they say.) ;)   
   >>   
   >>   
   >>   
   >> Cheers   
   >>   
   >>   
   > And you can measure the spectrum with that nifty Ocean   
   > spectrophotometer you have. What kind of output does it have? BTW, I   
   > remember your rotating a diffraction grating on a servo - Would it be   
   > feasible to make a "home brew" spectrometer or would this be a big   
   > reach?   
      
   It isn't all that difficult, except for making the fibre bundle.  Mine   
   was all built out of Microbench stuff from my drawer, plus some toy   
   parts.  (You can get servos with magnetic encoders now, and with a $40   
   programmer, you can reduce the size of the dead zone, which would have   
   helped the repeatability of mine.  A preload spring would have been a   
   good idea too.)   
      
   The other issue is keeping the stray light down, which takes a certain   
   healthy paranoia.  I used a lens instead of a mirror to collimate the   
   light on the grating side, and a near-Littrow configuration to save   
   alignment.   
      
   Most of the fine mechanical stuff (e.g. the fixtures for aligning the   
   fibre bundle) was made from JB Weld epoxy.  A fun project all round.   
      
   Cheers   
      
   Phil   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)