From: "Marvin Kaye" <marv@lancaironline.net>
Subject: Re: [LML] Re: Lancair 320  Fire 
To: lml
Date: Fri, 26 Jan 2007 18:31:41 -0500
Message-ID: <web-1792807@logan.com>
In-Reply-To: <8984A39879F2F5418251CBEEC9C689B32876C1@lucky.dts.local>
References: <8984A39879F2F5418251CBEEC9C689B32876C1@lucky.dts.local>
MIME-Version: 1.0
Content-Type: text/plain;charset="iso-8859-1";format="flowed"
Content-Transfer-Encoding: 8bit

Posted for "Chuck Jensen" <cjensen@dts9000.com>:

  I was curious if you'd considered the failure steps involved with
  something like a cracked exhaust header or fitting.  The break will
  start dumping hot gases into the cowl area (hence the advantage of
  thermocouples as fire detectors) but one is unlikely to see the wave
  length from an actual fire before the Uvtron will be triggered.  Do you
  want to let it progress all the way to a full blown fire before you are
  alerted?  Obviously, a fuel and some electrically driven fires will be
  visually apparent early on, but it seems like exhaust system failure,
  probably one of the leading causes of fires, would not be detected until
  you had a serious issue on your hands.  Just a thought.
  
  Chuck Jensen

"""
   I used a UVtron based fire detector designed by the robotics guys to
   help robots find fires in burning buildings.  I mounted it in a small
   project box and wired it to a big red light on my panel.  It only
   detects actual fire (flames), not light.  So, it won't trigger based
  on
   glowing exhaust headers, sunlight, etc.  Attached is a picture of the
   unit with the box open. The heart of the unit is the UVtron bulb,
  which
   is the clear bulb you can see mounted on the circuit board & standing
   upright within the box.  The "warning light" is the small LED lying on
   top of the box.  I ended up using a much larger LED than what is shown
   here.  You can assemble what you see in the picture for about $150.
   Hopefully, I'll never have to find out how well it works.
   
   Mark S.
   LNCE 208TX
"""