Hello John,
>I was very spurprised to see an
> enormous differnce in insertion loss as a function of the core diameter:
That should not be a surprise.
Core diameter has a LARGE effect on loss in transformers like this. The
magnetic path must be kept short, because loss tangent of high permeability
cores is high. Twisted windings or transmission line windings help do this.
<SNIP>
> CONCLUSION:
> Contrary to what you stated, in order to obtain the lowest attenuation,
> both the 43 and the 75 material cores needed twice as many turns as the
> MN8-CX cores.
That runs contrary to my experience, and the experiences of many others.
Motorola selects cores with a low frequency ui in the order of 125 (Q1
material) for HF amplifiers, and specifically discusses ui in their
"Wideband Impedance Matching" applications text for 3-200 MHz ranges, and
ui of 800 for .5 to 50 MHz applications. They suggest a low ui core like
125 to minimize loss in high power applications where a tiny bit of loss
can result in severe heat problems.
Directly quoting Fair-rite products, in an application bulletin called "Use
of Ferrites for Wide Band Transformers":
"Maganese Zinc ferrites, such as 77 or 73 materials, are very suitable for
low and medium frequency broadband transformer designs." For HF
transformers they suggest using 73, 43, 65 or 61 materials.
Siemens does the same, as do a total of 11 other technical books and
applications bulletins I have on broad-band transformers including "Soft
Ferrites Properties and Applications". The last text is considered the
"bible" of ferrite core engineering, selection, and transformer design.
This started out by my making what I considered a harmless and helpful
suggestion that 73, 43, or 77 mix cores would work and actually be a better
core. The basis of that statement was:
1.) I use such cores on daily or weekly basis in broadband transformers
for military, commercial and medical RF equipment and have spent many
hundreds of hours trying different cores and winding styles.
2.) I asked someone much more knowledgeable than I about ferrites, and he
agreed the mixes I suggested were much better choices for both loss and
number of turns required for a given core and winding style.
Rather than continue an obvious disagreement over core selection and/or
measurement methods that no one would benefit from, I'm willing to concede
(for the purpose of this discussion) my measurements, textbooks and
catalogs with published data on these materials, and engineers who deal
with such materials on a daily basis including those from companies like
Motorola, Siemens and Fair-rite products are wrong.
73 Tom
--
FAQ on WWW: http://www.contesting.com/topband.html
Submissions: topband@contesting.com
Administrative requests: topband-REQUEST@contesting.com
Problems: owner-topband@contesting.com
|