From: dhky@shaw.ca   
      
   On 05/09/2014 6:15 AM, Daniel wrote:   
   > On Saturday, 1 January 2000 19:00:00 UTC+11, D.H. Kelly  wrote:   
   >> Your concern over this assertion was well placed. It would have been   
   >> possible to use integral of H.dl around loop = current enclosed- hence H and   
   >> flux outside the core whatever the mu.  (which you did as part of your>   
   development.)  You went a step further and I, for one, am glad that you did.   
   >> Don Kelly   
   >> dkelly@nabunalimo.lark.com   
   >> remove the bull to reply   
   >>   
   >> BillyFish  wrote in message   
   >> news:20000101001804.15555.00000237@ng-ce1.aol.com...   
   >   
   >>> Let me point out that this whole thread started because someone made the   
   >>> assertion that increasing the mu of the core to very high values would   
   >> prevent   
   >>> any flux outside the core.  I think that the consensus now is that such an   
   >>> assertion is erroneous.  This glib assertion suckered me in.  In a sense,   
   >> it   
   >>> was a scam artist at work.  Well meaning maybe, but a scam artist   
   >> nevertheless.   
   >>>   I just had to *understand* why it was erroneous.  The scam misses the   
   >> point   
   >>> that a parallel H and E do not generate a Poynting vector.  Once that is   
   >>> realized, everything else follows rather easily.   
   >>>   
   >>> Bill   
   >   
   > so after 14 years was the conclusion reached that leakage inductance is in   
   fact essential to transformer power transfer.. or an unfortunate, usually   
   small but inescapable component which reduces coupling and hence reduces   
   efficiency?   
   >   
   > thx...   
   >   
   Diclaimer:- I am not the Don Kelly "quoted" above. I vaguely recall   
   someone making statements in my name.   
   If you removes the "bull" to reply- then you are replying to an empty   
   message.   
      
   Leakage reactance is NOT "essential" to transformer power transfer. It   
   is a result of the  fact of life that two conductors cannot be   
   superimposed to share the same space.  One can come close-particularly   
   with some toroidal LV transformers-but....   
      
   However, efficiency in the normal sense of   
             (real power out/real power in)   
     depends on resistance, not leakage reactance.   
      
   There is no wonderful conclusion after 14 years- the conclusion came   
   over 114 years ago. That is that coupling is not perfect-hence leakage   
   reactance.   
   --   
   Don Kelly   
   remove the cross to reply   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)