Re: Primary question / Terman

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Bert Hickman by way of Terry Fritz <twftesla-at-uswest-dot-net>" <bert.hickman-at-aquila-dot-net>

Dr. Duncan and all,

Here's a vote for Terman - he's half right! It depends on how you wind
your primary!

Terman is absolutely correct for the case of an "edge wound" coil made
from a flat conductor (such as a helix wound the "hard" way - see ASCII
art below). In this case skin and proximity effects will tend to
distribute conduction toward the thin edges as Terman suggests, and the
result is poorer than for a round conductor. However, the turns can be
spaced closer, and since the inductance increases as turns squared, you
may still end up with a coil having higher Q in the same volume.

 Terman is correct for a edge-wound helix:
=========================================

    ---         
                   ---
    ---
                   ---
    ---         
                   ---
    ---
                   ---
    ---         
                   ---

A good rule of thumb for both skin and proximity effects is that high
frequency currents will tend to flow in such a manner which minimizes
the inductance of the total path. So, without having to resort to heavy
math, it can be seen that a flat spiral TC primary will tend to have
current flowing preferentially along the broad flat surfaces (skin
effect), with slightly more current flowing on the surface facing the
secondary winding (proximity effect). So, at least for the configuration
below, a flat conductor is significantly better than a round one. 

Terman is not correct for this type of spiral coil:  
==================================================


        | | | | |          | | | | |
        | | | | |          | | | | |


Hope this helped!

-- Bert --
-- 
Bert Hickman
Stoneridge Engineering
Email:    bert.hickman-at-aquila-dot-net
Web Site: http://www.teslamania-dot-com

Tesla list wrote:
> 
> Original poster: "Dr. Duncan Cadd by way of Terry Fritz
<twftesla-at-uswest-dot-net>" <dunckx-at-freeuk-dot-com>
> 
> Subject: Primary question
> 
> >Original poster: "Chris Brick by way of Terry Fritz <twftesla-at-uswest-dot-net>"
> <cbrick-at-rebelbase-dot-com>
> >
> 
> Hi Chris!
> 
<SNIP>
> One thing puzzles me about strap primaries.  Dear old Frederick Emmons
Terman,
> ScD, bless him, says that flat strip conductors are very poor at radio
> frequencies because the current travels along the two edges ("Radio
> Engineering"
> p20) where it is constrained by the magnetic flux, and he uses this to
> illustrate the point that it isn't the surface area of the conductor which
> matters.  Everyone else I have read says (sometimes with heaps of partial
> derivatives and Bessel functions) that the current travels on the surface and
> hence flat strip is very good.
> 
> [Pained hush prior to near-heresy . . .]  Could Terman be wrong?  Or have the
> mathematical whizzos missed some fundamental physical reality hidden in
amongst
> the mathematical mud pies?  I'd like to believe Terman and I can see his
> logic -
> if you take a cross-section of a circular wire, the current flows along a
thin
> ring at the circumference (skin effect) so if you take a slice across the
> diameter, the current will be along the two opposite edges.
> 
> At what density of current flow, frequency, thickness, width and
resistivity of
> strap does the magnetic flux no longer hold the current at bay at the two
> edges?
> It will be connected with the penetration of the flux into the strip, and the
> linkage of that flux with the current I know.  A.G. Warren (chapter 17,
"Bessel
> Functions" in "Mathematics Applied to Electrical Engineering", Chapman &
Hall,
> fifth impression 1946, volume 9 in the "Monographs on Electrical Engineering"
> series, editor H.P. Young) effectively says Terman is in error, but he
assumes
> the current to flow on the faces of the strip and then proves it, if you see
> what I mean, and it seems to me that his model could apply just as well
if the
> current was flowing at the edges.  Answers on a hyperbolic function please. .
> 
> Chris, just go ahead and do it.  Never mind the academic pedantry, my brass
> strip primary worked just fine.
> 
> Dunckx