[RFI] shielding answer

Jim Brown jimbrown.enteract at rcn.com
Wed Dec 24 16:05:15 EST 2003


On Wed, 24 Dec 2003 16:23:28 -0500, Tom Rauch wrote:

>When a time-varying electric field is taken to zero, so it the magnetic
>field. When a time-varying magnetic field is taken to zero, so is the
>electric field.
>
>The idea we can "shield" or filter one field and not the other is the basis
>of all sorts of antenna hokey-pokey and voo-doo, like the CFA or E-H antenna
>or "magnetic loops" that only respond to magnetic fields and thus somehow
>magically filter noise from signals that are on the same frequency.

Not really. Consider the near field of a 1/4 wave resonant antenna.
Near the feedpoint, the magnetic field will be quite strong, but the
electric field will be quite small. At the far end, the converse will
be true. 

Consider your proposed test fixture. There will be inductive coupling
between the exciting and three conductors running parallel to it (i.e.,
shield, red, and black), and a perfect shield won't do much to change
the inductive coupling to red and black. On the other hand, the
grounded shield greatly reduces the capacitive (electric field)
coupling between the exciting conductor and the red and black
conductors.  

Many serious cases of EMI happen in the near field of the source. Thus,
the relative importance of the E-field and the H-field IS a matter of
interest, and is far from voo-doo. If the H-field is dominant, a cable
shield won't do much, but twisting will. Conversely, if the E-field is
dominant, twisting helps, but the shield can be quite effective. 

Jim Brown K9YC



More information about the RFI mailing list