Re: eddy current with secondary coil

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

Original poster: "Antonio Carlos M. de Queiroz" <acmq-at-compuland-dot-com.br> 

Tesla list wrote:

 > Original poster: Ed Phillips <evp-at-pacbell-dot-net>

 >         By classical filters do you mean such things as the horrific
 > "m-derived" design or just passive filters in general?

The horrible m-derived, and the equally terrible, but at least
theoretically exact, doubly terminated designs that replaced them.
Doubly terminated LC filters with maximum power transfer are
quite insensitive to errors, but even so the element values must be
at a few % of the ideal design for an accurate filter. The sensitivity
problem is worse in active filters, but at least there are no inductors.

 > I've noticed
 > that guys who do filter design for a living often bridge their
 > capacitors to be as close as practical and, when really tight tolerances
 > are required, adjust the associated inductors to some criterion such as
 > resonant frequency.  For production filters a part of the design may be
 > a detailed alignment procedure for each circuit branch which could be
 > isolated.

Yes, tuning procedures can be devised, by adjusting resonances with
parts of the filter short-circuited. But it's simpler to have the
parts precise from the start.

 >         The design of "modern RF microelectronics" has certainly promulgated
 > the use of extremely sophisticated modeling tools which are probably
 > overkill for normal inductors.

But it's interesting to look at papers on inductor modeling, some
over 100 years old, where the claimed precision of the formulas is
in parts per million.

 >         I haven't tried to build a magnifier but can see your 
point.  When you
 > say "it's difficult to adjust the inductors later" I assume you mean
 > that it's hard to correlate the tuning and its results?

For a design aimed at very fast energy transfer, changing a primary tap
point changes the coupling coefficient between L1 and L2, and both are
critical. Ok that the coupling can be adjusted too, but then the
capacitance between L1 and L2 changes, and it appears in the
design and shall be adjusted too. It's actually not difficult to tune
the system while observing waveforms at low power, but with everything
calculated with precision, it's just a question of assembling the
system and it works as predicted (if the insulation holds...).

Antonio Carlos M. de Queiroz