[Amps] RF Chokes

R.Measures r at somis.org
Tue Feb 1 14:22:03 EST 2005


On Feb 1, 2005, at 4:41 AM, Angel Vilaseca wrote:

> Hello Richard
>
> Thanks for your answer. I do not have a professional background in
> electronics so from time to time I need to ask some questions that will
> probably seem quite basic to many of this list readers. Please correct
> me if you find errors in what follows.
>
>> From what I understood reading the literature I could find, core
> saturation is related to the current in the windings. Any
> core can saturate, even at audio frequencies, even at DC, if you apply
> enough
> current and if you use few windings.

Saturation is related to ampere-turns and to frequency.  At very low 
ampere-turns, Mu = 125 material can go well above its 10MHz rating 
without saturating or heating

> Indeed, for a given signal power,
> with increasing frequency, reactance would also increase hence current
> in the windings would decrease and the core would go away from
> saturation. So how could a core saturate *above* a given frequency?
> On the other hand, if you use a core at too high a frequency, the core
> will heat due to RF losses, not saturation.

So it would seem, however the increase in XL with frequency is not 
enough to prevent saturation and distortion of the waveform -- i. e., 
harmonic generation.  In other words, the problem is not just core 
heating, it is RFI.
>
>
> This is a different effect.
> Unlike saturation, RF heating will increase with frequency and happen 
> at
> any power level, even very small.

Can a small amount of power cause a heat problem?
>
> Now for the harmonics generation:
>> From what I could understand from the handbooks, to produce 
>> harmonics, a
> given device must have non-linear properties. You can use a class-c
> biased transistor or tube or a diode for instance.
>
> How severe is harmonics generation in a ferrite?

At this QTH, at 1400w-pep, a ferrite-core, 1 to 1 balun for an 
"Ultimate Transmatch" operating in the  7MHz band  produced enough 
harmonic energy to Ultimately wipe out an S9 signal from a line-of-site 
TV station on 555MHz.

> A ferrite core going
> from saturated to non-saturated will obviously behave non-linearly and
> generate harmonics. But the transition from linear to non-linear needs
> to be sharp for an afficient harmonics generation.

At 7MHz, there are 14-million sharp transitions every second.

> It is indeed very
> sharp in a snap-varactor, for instance, but how sharp is it in a
> ferrite?

It happens.  This is why ferrite manufacturers rate the max freq of 
various mixes.  For instance, Mu = 125 material is rated at 10MHz.
>
> I did not have much luck trying to find information about harmonics
> generated by ferrite cores, searching  with Google or in the handbooks.

There is no substitute for finding out hands-on with an oscilloscope 
and 1500w.
>
> As a bottom line, I would tend to think that the best protection 
> against
> ferrite-generated harmonics is to use as large a core as practical, 
> with
> low permeability, and with a lot of windings over it.
>
> What do you think?

Whatever does not produce distortion of the sinewave on an oscilloscope 
is the bottom-line.  Powdered-iron is much more resistant to harmonic 
distortion than ferrite.
>
>
> Regards
>
> Angel Vilaseca HB9SLV
>
> "R. Measures" wrote:
>
>>>
>>> If a ferrite core is used, could the ferrite saturate and generate
>>> harmonics?
>>
>> Yes.  Mu = 125 ferrite rod material begins to saturate at c. 13MHz, so
>> it's a risky business.  Powdered-iron has the advantage of not
>> saturating.
>>>
>>> How large sould the ferrite core be and which mu should it have?
>>
>> For less than 13MHz, Mu = 125 is okay.
>>>
>
>
>

Richard L. Measures, AG6K, 805.386.3734.  www.somis.org



Richard L. Measures, AG6K, 805.386.3734.  www.somis.org 



More information about the Amps mailing list