RE: 2 layer primary puzzle

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

Original poster: "Rothman, Aric by way of Terry Fritz <twftesla-at-qwest-dot-net>" <Aric.Rothman-at-Honeywell-dot-com>

> -----Original Message-----
> From: Tesla list [mailto:tesla-at-pupman-dot-com]
> Sent: Tuesday, September 25, 2001 3:27 PM
> To: tesla-at-pupman-dot-com
> Subject: 2 layer primary puzzle
> 
> 
> Original poster: "James T by way of Terry Fritz 
> <twftesla-at-qwest-dot-net>" <jamest2000-at-att-dot-net>
> 
> Hi Everyone,
>  I will keep this brief. I am working on an 8" coil with a 2 
> layer primary.
> I have constructed the primary and
> it is not behaving as I expected. I measured the inductance 
> of the primary
> by itself.
>   What happens is, I measure from the outside of the top coil 
> to the inside
> and the inductance goes up as
> expected. Then I drop down to the bottom layer and measure from the
> top/outside to the bottom/turn, oops, the
> inductance goes down as I go out. The inductance of the 
> bottom coil seems
> to cancel the inductance of the top
> coil. Please set me straight on this.
> 1.understanding/myth - the winding direction between the 2 
> coils is irrelevant.

The winding directions are relevent if k > 0 (k = coupling coefficient).
Ideally, two perfectly coupled inductors of equal value (L) with opposite
sense (wound in opposite directions) will appear as a wire (zero
inductance).  If k=0, then the inductance will be 1/2 L.  As k -> 1,
inductance will approach zero for windings of opposite sense, and will
approach L for windings of identical sense.

> 2. understanding/myth - the "mutual inductance" of the 2 
> coils increases
> inductance, not cancels out the
> others inductance.

In the case of two paralleled inductors, with variable coupling between
them, mutual inductance will not increase the inductance beyond L (the value
of our two hypothetical, equally valued inductors).  When k=1, it is as if
there is only one inductor of value L.  When k=0, you have two like valued,
uncoupled inductors in parallel, in which case Ltot = 1/2 L (uncoupled
inductors behave like resistors, with inductance taking place of resistance,
when forming nets).

> 3. 2 layer primary's are a practical approach. I seem to 
> recall some posts
> on this, and it was considered
> reasonable. Brilliant to ask now that it is done!
> Any comments will help. Thanks,
>  James Cart

Others have had sucess with coupled, paralleled primaries.  Whether or not
you have the primaries in opposition depends on the primary inductance you
will need to achieve resonance.  If we call the inductances of the two
co-primaries to be L, and the inductance needed to achieve resonance Lr, you
will use opposing sense on the primaries for Lr < 1/2 L, and same-sense for
Lr > 1/2 L.  Be mindful that if the opposing sense route is taken, the
currents in your tank circuit will become very very high as Lp -> 0.  For
this reason, you would be better off using smaller co-primaries, oriented
the same way turn-wise, if you find Lr < 1/2 L for your current
co-primaries.

BTW: I hope other will either reinforce the above, or quickly point out I am
full of beans, as I am writing this off the cuff, without bothering to do
any calculations.  I encourage you to flip one of the primaries and measure
the inductances at varying separations to verify what I have said here.

Aric