Hi Dave,
I've studied ferrite chokes extensively since 2003, published landmark
work on it as an AES Paper in 2005. There's a lot of work on my website
about it, k9yc.com/publish.htm My first publication of that work in the
ham world was in 2007.
Some things I showed is that resistance is more useful than inductance,
and capacitance is GOOD because it allows us to put the resistance
formed by the low Q resonance where we want it by winding the "right"
number of turns.
#31 is a FAR superior material at HF because it has both dimensional and
circuit resonances, which combine much as the old stagger-turned IFs to
increase the useful bandwidth. #43 is a terrible material for HF,
because of the broad manufacturing tolerances move that resonance around
-- it's inherently higher Q because it's NiZn. #31 is a very special
MnZn mix, which gives it the dimensional resonance, and the broader
resonance reduces the problem with manufacturing tolerances when you
want to build something in quantity or publish a set of designs, as I
did four years ago. I spent three years on that HF choke project. My
process is described in the Cookbook.
On my website, check out my Ham's Guide to RFI, which develops the
resonance concept, as well as a lot of other stuff. And the HF Choke
Cookbook, which includes those designs and tells the engineering project
it took to develop them. And the VHF/UHF cookbook which describes
designs like the one you are talked about here, except for coax and
two-wire line. And when I've designed chokes for the AC line (to cool
off noisy generators, for example), I wind three-wire jacketed cable or
Romex through the cores. There one for HF in the Handbook, and on page
23 of http://k9yc.com/RFI-Ham.pdf
And I sure agree with you about manufacturing screwing up a good design.
In the short time I spent in mfg, it was the department whose job it was
to build it cheaper. This was when we still manufactured in this country.
73, Jim K9YC
On 2/18/2022 3:19 PM, David Eckhardt wrote:
An effective common mode choke on the mains conductors should pass all
three conductors through the core. For a 2.4-inch OD core, three turns
are usually good. Too many more will increase the capacitive coupling.
31 material has an advantage at the lower frequencies where the older 43
material is a bit better at higher frequencies. All referring to HF.
I have "solved" many an emissions problem usually around 40 to 70 MHz
from SMPSs by simply installing a simple common mode choke on the mains
lines (yes, all three lines: hot, neutral, and "ground" - the green
wire). Once these products reach a contract manufacturer, the common
mode choke adds nothing to performance and only adds cost, so it is
removed. Many times I ran into the reality that contract manufacturers
are the EMC engineer's worst enemy. They are supplsed to retest and
re-qualify, but they don't.
_______________________________________________
RFI mailing list
RFI@contesting.com
http://lists.contesting.com/mailman/listinfo/rfi
|