hello, Tom
>
>While I agree an air core alone is sufficient to make a reasonable
>filament choke on 50 MHz, Rich has a total misunderstanding of
>what creates "distortion" that we can hear in a linear PA.
>
It was not a SSB signal. It was an AØ signal of 100w. The type-61
material performed well at 10MHz. The waveform looked good. I slowly
increased freq. At 14MHz, the waveform did not look good. . However,
with a ferrite material rated for 50MHz max. instead of 10MHz max., no
distortion was observed at 14MHz.
>Not only that, something must have been wrong in Rich's test
>setup. Operation at higher frequencies does not suddenly create
>waveform distortion in a ferrite material.
>
Sweeping from 10 to 14 MHz did not produce any sudden change. The change
was pretty gradual.
>If a certain material did create harmonic distortion as Rich wrongly
>claims, the manufacturer would NOT recommend it for EMI
>suppression!! That's because it would create harmonics and create
>EMI through generation of harmonics.
>
Perhaps this is why we often see a bypass cap on the outgoing side of a
ferrite RF attenuator bead?
>The claim operating 61 material at 14 MHz creates harmonics is
>nonsense. Clipping is a flux density problem, not a frequency
>problem.
The observed waveform was not clipped at 14MHz. . Between 80-deg. and
100-deg., the waveform looked pretty much like a sine wave. // Permag
Pacific clearly indicates that the permitted flux density varies with
frequency. I suspect that if I had decreased P, the waveform at 14MHz
might have improved.
>>
>> ? In the test I performed with type-61 material operating at 14MHz, the
>> slope of the waveform was noticably different from that of a sinewave --
>> esp. between 60-degrees to 80-degrees and 100-degrees to 130-degrees. .
>> My guess is that the distortion was >15%.
>
>Great "guess". What scope did you use? Maybe it needs repair.
>
? A Hewlett Packard 1807. The RF from my 2m HT produces a sinewave on
the screen.
>> >Quoting from Fair-rite Corp, an actual manufacturer of soft iron
>> >materials, 61 material is suitable in broadband transformers and
>> >EMI/RFI suppression devices up to 200 MHz, and tuned
>> >applications up to 10 MHz. Q at ten MHz is 80.
>> >
>> ? The cathode of a g-g amplifier is akin to a tuned circuit.
>
>Nonsense. The choke is intentionally designed so its impedance is
>several times the cathode impedance. That means the choke has
>little RF current, low magnetization, and is NOT a part of any
>resonant circuit.
>
>Besides that, the eddy currents (the "thing" that limits upper
>frequency Q) simply add linear loss resistance to the system.
>
>> >> I would consider a filament choke to be in the latter category and if
>> >> its inductive reactance is the correct value, its impedance should be
>> >> far greater than that of the cathode and therefore not affect the
>> >> waveshape a significant amount That effect has NOTHING to do with
waveshape
>> >> distortion.
>> >
>> ? Harmonic distortion has everything to do with sine waveshape
>> distortion.
>
>That's right. But harmonic distortion has no ill effects in a narrow
>band PA.
>
It does when your neighbors are watching the NBA playoffs.
>People who aren't totally familiar with PA operation sometimes
>confuse harmonic distortion with the type of distortion that causes
>splatter. Splatter is NOT caused by harmonic distortion, splatter is
>caused by amplitude non-linearity in the envelope shape.
>
Is this a plea-bargain?
>That's the reason a single-ended class B PA is just as linear and
>can produce just as low distortion in a SSB signal as a push-pull
>PA.
>
¿say what?
>If 61 material did cause harmonic distortion (which it doesn't), it
>could STILL be successfully used in a resonant PA.
Agreed - provided the flux-density was sufficiently reduced above 10MHz.
However, a material with a Mu of 40 and a max. freq. rating of 50MHz is
available at the same cost, so why not use it?
>It would be useless in broadband applications, and EMI suppression.
>
>There is no other explanation for your results and conclusions.
One explanation is that someone made a somewhat faulty assumption.
>
>"The test" you made must have been flawed, and your end-conclusions
>make absolutely no sense at all when compared to well published
>data.
The results of the "The test" seemed to be in general agreement with
Permag Pacific's little orange book.
>
- cheers, Tom
Rich...
R. L. Measures, 805-386-3734, AG6K, www.vcnet.com/measures
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|