[RFI] LSI

[Jim Brown]

On Fri, 9 Jan 2004 08:30:06 -0700, Richard Zalewski wrote:

>I was able to solve the problem
>on 10m by putting ferrite on the power cord of the control unit but still
>causing a "problem" on 40M.  More ferrite?

As Tom, W8JI, noted recently in an excellent post, there's only so much
you can do with ferrites based on their impedances and impedances of
the circuits into which they are inserted at the frequency of the
interference. That said, a ferrite material designed for use at a lower
frequency, and/or winding more turns around a ferrite core, may solve
the problem.  The excellent applications notes that are part of
Fair-Rite's pdf catalog, as well as a study of their data sheets for
various products, may lead you to a solution. This catalog is a
must-have for anyone trying to understand how ferrites work. 

Example of one of the things I learned there -- the impedance peaks of
a ferrite on a conductor will shift lower in frequency as the number of
turns around the ferrite is increased. No surprise when you think about
it -- stray capacitance has a lot to do with it. 

It is generally well known that the equivalent circuit of a ferrite
around a conductor is an inductance in series with a resistance. At low
frequencies, the inductance dominates. At some high frequency
(different for each material), the material becomes far less inductive
but much more lossy. 

Add this to the well known fact that the inductance (and thus the
impedance) at the lower end of the useful frequency range of the
material  will increase as the square of the number of turns, and it's
clear that more turns may be the answer. 

Jim Brown K9YC