> Before you make this statement, you should measure chokes
built with some of
> these cores. I have done so, and in a disciplined way.
I already have Jim. I just completed measurements a few
weeks ago on 31 material using identical size samples to my
standards of other materials.
I always do this as an approval process.
I can also read catalogs, and FairRite corps catalog shows
31 has a Q=1 at about 3MHz. 44 material, the most common now
for snap-on beads from many sources, has Q=1 at 20MHz. 43
material has Q=1 at 10MHz, and has noticeably less impedance
per unit length of core against the conductor.
All of those materials are the most common. 31 and 44
actually have similar performance at upper HF, but 31 has a
slight edge especially down low (around 1.8 MHz). 31,
however, is a newer material and not well stocked.
In contrast, 73, 75, and 76 materials have Q=1 at 2, 0.7,
and 0.2MHz respectfully with 73 or 75 material probably
providing the best RFI suppression of any material over the
HF region for a give core material length.
My points are that:
1.) Throwing a few beads (even with multiple turns) at a
problem and "assuming" the world is filled with devices that
have modest or low common mode impedances so the beads
actually work is not really a good RFI-proofing protocol. It
is one that many people use, and they need to be educated as
to how poorly this method actually works.
2.) People also assume a distributor's suggestion (like
Amidon) for "frequency range" are good, but they almost
always are NOT. For HF, the best performance generally comes
from 73 material. You want an impedance peak at HF, and
hopefully with very low Q over a very wide bandwidth.
3.) Snap on cores are really are not all that good at best,
and the worse ones really are next to doing nothing.
4.) A few small bypass capacitors and series chokes will do
significantly more and do it much cheaper. If you don't want
to cut wires at LEAST install some bypass caps or ground
unused pairs in the cable to add capacitance to the
equipment ground with your beads!!
5.) The type of cable or number of twists per inch has
little to do with RF ingress from sources more than a few
dozen conductor spacings away.
The single most effective thing anyone can do is make
everything in the room around the affected device have the
same RF potential.
73 Tom
_______________________________________________
RFI mailing list
RFI@contesting.com
http://lists.contesting.com/mailman/listinfo/rfi
|