From: nobody@nospam.not   
      
   Morris Dovey  wrote in   
   news:js22m4$3gp$1@speranza.aioe.org:   
      
   > On 6/22/12 7:45 AM, Han wrote:   
   >> Morris Dovey  wrote in news:js0eu2$e0p$1   
   >> @speranza.aioe.org:   
   >>   
   >>> I'm thinking that I want to establish an average background count as   
   >>> part of the start-up procedure, then trigger an automatic shutdown   
   >>> if the unshielded count exceeds that average by some factor (x10?)   
   >>> or by a lesser factor (x5?) through a 30-second period during the   
   >>> test.   
   >>   
   >> "Counts" have a statistical fluctuation.  If you have a level of   
   >> counts 10- fold over average background (averaged over a few seconds,   
   >> up to a minute), it is really a very minimal amount.  To exclude   
   >> false positive shutdowns, I'd start with at least 100-fold over   
   >> background, averaged over a minute.   
   >   
   > The change to average over the past 60 seconds was quick and easy, and   
   > I'll increase the safety thresholds to split the difference - but   
   > until I have some data to work with (translation: "confidence") I'll   
   > live with the false shutdowns, and adjust to suit my level of   
   > frustration with interrupted tests. :-)   
      
   I really do understand the better safe than sorry philosophy.  OTOH, many   
   people live in (relatively) high radiation environments (subways, high   
   altitudes, frequent air plane flights).  Therefore the background   
   radiation level in your neck of the woods is probably really low.   
      
   --   
   Best regards   
   Han   
   email address is invalid   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)