I have been experimenting with ferrite loopsticks over the past couple of
months It appears that the diameter of the loopstick itself is a major
factor in determining the output level of the antenna. Length doesn't seem
to have a major effect.
Rod material with a permeability of 125, such as Type 61 material, is
specified to work from 200kHz to 10MHz. In actuallity, the material can work
down to near DC, and still retain its properties.
When a material is specified over a frequency range, we are talking about a
one-side spec. You can always use the material over a range lower in
frequency than its specs call for, but not above that frequency. I don't
know if this is because of magnetic hysteresis losses at higher frequencies
or not.
There is an initial permeability spec, which is the number that we look at,
and there is an effective permeability, which is the real world number. I am
not sure if this occurs because of external factors, such as the effect that
the wire wound on the ferrite rod has.
The greater the aspect ration, the closer the effective permeability will be
to the initial permeability. This is a good thing.
Aspect ratio is defined as Length/Diameter of the rod material itself.
I have used 8 inch long rod material that is 3/8 inch in diameter and I have
also used 6 inch long rod material that is 5/8 inch in diameter. The 6 inch
5/8 inch diameter material has higher RF output, and the nulls seem to be
just as deep as the 8 inch diameter material.
Since I am using a differential JFET amplifier chain, part of the design
parameters include matching the JFETs for gain. This gives you the best
common mode rejection from electrical noise, and it also results in deeper
nulls. If you order a couple thousand JFETs from the same batch, the
characterstics of these devices will be closely matched.
The best approach would be to use something like a Siliconix E431, which is
a dual JFET in a single package. This would provide very close matching,
since both devices are on the same silicon die.
The ferrite information that I provides is courtesy of Ray, at National
Magnetics, except my conjecture about reasons for initial permeability vs
effective permeability.

Pete