From: don@manx.misty.com   
      
   In , Salmon Egg   
   wrote:   
      
   >A friend of mine claims that he had a Primus lantern with mu h greater   
   >light output than a comparable Coleman lantern. Both used derivatives of   
   >the Welsbach mantle.   
   >   
   >Does anyone know of light measurements comparing such lanterns or   
   >mantles?   
   >   
   >Is there any indication that one brand of mantle works better than   
   >another one does?   
   >   
   >Is one brand of lantern better able to heat its mantle to a higher   
   >temperature than another?   
   >   
   >Are there any modern investigations of how mantles work?   
   >   
   >Is a luminescent process of any importance?   
   >   
   >Can selective radiation explain everything.   
      
     I think selective radiation is *THE* explanation.  The selective   
   radiation is also temperature dependent, but temperature dependence of   
   the selective radiation is not necessary for selective radiation to be   
   successful in gas mantles.   
      
     It appears to me that gas mantles have something where the selective   
   radiation is from a band of upper states to a band of lower states, or   
   from a "broadened upper state" to a "broadened lower state",   
     involving at least 1 of these 4 items.   
      
     The lower state appears to me to be above "ground state".  Ordinary   
   flame temperature gets the material in question changing from "whitebody"   
   to "selective radiator".   
      
     A non-selective radiator radiates IR so much that it won't get as hot as   
   a selective radiator can.   
      
     For example, fine metal wires appear to me to get heated to only about   
   1500-1600 K by propane torches.  A selective radiator can probably   
   achieve 1800 K, maybe 1850 K, in a propane torch - or 1900 K in one with   
   unusually lean fuel/air ratio that approaches complete fuel oxidation   
   independently of outside-of-the-flame air.  ("Richer" than combustion   
   nearly complete available from pre-mixed air is necessary to avoid   
   oxidizing most metal brazing and soldering targets.)   
      
     Improving radiator temperature from 1600 to 1900 K increases luminance   
   by a factor of about 12.  Improving radiator temperature from 1500 to   
   1800 K increases luminance by a factor of about 15.   
      
     And how hot can a mantle made of blackbody threads over a wide area   
   get from a slower-flame-gas-speed get?  Possibly only 1400 K?  Improvement   
   in luminance is by a factor of 32.   
     Even if the selective radiator avoids radiating visible red wavelengths   
   enough to make its luminance half that of a blackbody at its temperature,   
   at this rate this means luminance improvement by a factor of 16.   
     This is with allowing an explanation why CCT gets into the 3,000's   
   along with greener-than-blackbody color.   
      
     As for temperature necessary in a usual gas mantle, roughly a 2 cm   
   diameter 3 cm tube capped by a 1 cm radius hemisphere:   
      
     "Frontal area" sideways is nearly 7.6 cm^2; I would make 7 cm^2 due   
   to pointier-than-hemispherical tip and any overestimation of size.   
      
     Get that to be as bright as a 450 lumen 40-watt 120V lightbulb for   
   viewing or being illuminated by "broadside", and I think that's plenty   
   bright for camping at night.   
      
     Divide 450 lumens by 4-pi "at this rate", this means 36 candela from   
   7 square cm can do the job.  5.14 candela per square cm can do that.   
      
     Guessing mantle "direct frontal area" is 40% holes, 60% mantle, and   
   mantle "brightness emissivity" of 50%, at that rate the necessary   
   temperature is that which has a blackbody achieving 17.1 candela/cm^2.   
     That is 1850 K, which I think a selective radiator can somewhat easily   
   be heated to by a propane flame.   
   --   
    - Don Klipstein (don@misty.com)   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)