The 'big boys' who design the materials use one turn through the core to
characterize the ferrite material. They measure in a 50-Ohm system.
Remember, one turn equates to passing a conductor through the core, NOT one
turn *around* the core!
I have recently characterized most of my larger cores using this method.
It works nicely to compare the measured impedance curves (linear magnitude)
against the published data curves for each core material. The measured data
matched the published data well enough to determine the core material. I
built a fixture to accept the one-turn (based on the single pass through
the cores) to accomodate serial testing of my cores. The measured data was
made in the transmission mode so a complete cal. is required. I even did
isolation, just because I could. The measured data matched the published
data curves up to 200 MHzrelatively well. I didn't calibrate or measure
above that frequency as I was interested only in HF through 50 MHz. Of
course, the fixture was calibrated into the measurement setup.
This is also a good method for detecting resonances (to be avoided!) in
home brew CMCs and current 'baluns'. Getting to the data I put out earlier
today, many designs showed resonances which nixed the design of that
specific CMC. That which won consisted of 11 bifilar turns (no twists) of
AWG #10 stranded and insulated copper conductor wound on two stacked 3" OD
43 material cores. It measured a nice smooth convex curve from 1 MHz
through 50 MHz with no resonances in that range. That is the choke noted
as (BRN) in my data compilation as that was the color of the wire (I'll
attach the data again just for reference). However, the 31 material still
won out on 160-meters showing no resonances in the frequency range of
concern but was not as good at mid-HF frequencies.
Dave - W?LEV
On Thu, Dec 31, 2020 at 3:26 PM Manfred Mornhinweg <manfred@...>
wrote:
But if the choke balun is highly inductive and not very resistive, it will
form a resonant circuit with the antenna's total capacitance to ground, and
other reactances in the antenna/feedline/ground circuit. If this resonance
happens to fall near your operating frequency, the balun will make matters
worse then they are without a balun!
For that reason many authors suggest to use lossy core materials,
including authors already mentioned in this thread. Ferrite has a complex
permeability. At low frequency it's almost purely inductive, while at
higher frequencies it becomes increasingly resistive. For that reason, at a
sufficiently high frequency for the material chosen, a choke balun will
oppose little inductance, but a lot of resistance to any common-mode
current. This makes it work very well as a balun, without any risk for the
mentioned resonances in that frequency range. But its resistance must be
high enough to keep the loss negligible, or at least acceptable.
This comes down to selecting a core material that has a mostly resistive
permeability all over the intended operating frequency range, and winding
enough turns on it to keep the flux density at a level at which the loss in
the core is low. At the same time there shouldn't be so many turns that
inter-turn capacitance becomes a problem.
I have found the NanoVNA to be a great tool for measuring the
characteristics of ferrite cores! You can take an unknown core, wind one or
two turns of wire on it, or several if it's a small core, hook it up to the
NanoVNA, and see the resulting series RLC graph over the desired frequency
range. This gives you an instant idea of the core material's inductive and
resistive permeability curves! By comparing to various
manufacturer-provided data, you can make a good guess which material your
core is made from.
--
*Dave - W?LEV*
*Just Let Darwin Work*
