XPost: alt.lasers   
   From: jappel@linux01.gwdg.de   
      
   Hello Phil,   
      
   >>> One of the things I'm doing for them is developing highly stable lasers   
   >>> for, *ahem*, hostile and size-constrained environments.  I have a few   
   >>> books on laser locking, e.g. Ohtsu, but I really need a more recent   
   >>> summary of the field so I don't miss anything important.   
      
      
   > If I had a library or laser conference right handy, that would be great.   
   > ;)   
   >   
   > I don't need someone to do it for me. I just need to make sure I know   
   > what relevant things other folks have done, to double-check that my   
   > approach is the right one for the job.  For instance, although I'm using   
   > R-T locking with an optically-contacted ULE glass etalon, it might be   
   > that there's an approach using molecular absorption lines that would   
   > work better or be cheaper overall.  A bunch of things like that have   
   > been tried, and my most recent reference is over a decade old.   
      
   As far as I know Pound-Drever-Hall-locking to an ULE cavity in vacuum is   
   still the state-of-the art technique to lock away frequency noise at high   
   sideband frequencies down to the few Hz-range. It's main advantage is that   
   it gives a great signal-to noise ratio and therefore needs little averaging   
   time. That makes it appropriate for fast feedback loops. For higher sideband   
   frequencies than your feedback can achieve (for example there is a nasty   
   180° phase shift in the transfer function for frequency modulation via diode   
   injection current somewhere between 1-10 MHz for most diodes) it might make   
   sense to protocol your error signal and correct your measurement instead, if   
   that is possible.   
      
   In optical atomic clock applications very low sideband frequency phase noise   
   (<-> frequency drifts) matters too, and thus the next step then is to lock   
   your stable laser to a frequency comb and from there to a frequency chain of   
   oscillators with lower and lower sideband frequency noise - the last and   
   slowest step is the lock to an atomic transition.   
      
   If the requirements are not so high, or if the power consumption or the   
   budget don't allow this, finding some kind of an atomic transition line   
   close to your laser frequency indeed gives an alternative way to avoid long   
   term drifts of your resonator.   
      
   Among the newer approaches that I have heard about to replace the medium- to   
   high phase-noise sideband frequency locks by   
    * Ultra-High-Q glass resonators (such as bottle-resonators or disk-   
   resonators made out of single crystals or glas).   
    * Temperature-stabilized fiber interferometers.   
      
   Unfortunately I don't have references at hand right now, but maybe I can   
   find some during the week.   
      
   It really depends what kind of measurements you are going to do, to what   
   kind of phase/frequency noise you are sensitive to, but of course you also   
   know that.   
      
   All the best,   
   	Jürgen   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)