Hi Mauri,
> A plate choke must ?resonate? well above the operating frequency when
> installed and connected to the amplifier circuits.
The resonance you are talking about is the series resonance
which shows itself as an impedance minimum.
Of course, it must be avoided.
Below it sits the first parallel resonance,
which shows itself as an impedance maximum.
The usual tuning is to position the parallel resonance
at approximately geometric mean frequency of the band.
Please look at impedance sweep on http://www.geocities.com/va3ttn
You can see parallel resonances at approximately 35, 60 and 85 MHz.
Series resonances are at 0, 49, 73 and 93 MHz.
> Again, the mechanism of an antenna choke it's not that of a transmission line.
> Behaviour of a transmission line (when not resistively terminated into it?s
> own
> characteristic impedance) is to cyclically show high and low impedances, and
> repeatedly
> change in reactance sign, with an EXACT harmonical relation.
Every piece of wire IS a transmission line by definition.
It doesn't have to meet any particular criteria to be so.
The special case of a SHORT wire is conveniently regarded
as a "lumped" element (R, L, C). Wires used in chokes and
baluns are usually not short enough to be regarded as
lumped elements.
> We can agree that not only capacitance between turns has to be taken in
> account.
> Expecially when it?s small, sourrounding objects may be dominating as the
> parasitic
> capacitances.
A choke interacting with environment in ways other than via terminals
is an analysis nightmare :-) It's not a one-port device anymore.
Unfortunately, real baluns, especially those without ferrite,
are large enough to interact "wildly" with the environment.
> Circuit resonance anyway occurs when the reactances are equal and opposite in
> sign,
> and you can't find out more than a single resonance in an antenna choke
> between its
> terminals
That would certainly be true if a real choke was a "circuit".
However, "circuit" is an abstraction, as explained above,
and a real choke is too large to be a "circuit".
Multiple resonances that you can see on the sweep mentioned above
clearly illustrate that it IS a transmission line.
Of course, below ~25 MHz it can be reasonably regarded
as a "circuit", just as any other shorted transmission line
can be regarded as a "circuit" ("inductor") on low enough frequencies.
Maybe you can perform your own experiment?
Take ~6 m of 1 mm wire, make a "coil", and measure its impedance
from 10 to 100 MHz. You'll get more resonances that you ever wanted :-)
How do you actually measure choke impedance?
I use HP4193A for 1-100 MHz, HP4192A for LF, and E8358A
network analyzer above 100 MHz. The picture on the site was
measured with YT6A's Advantest R3261 spectrum analyzer / tracking generator.
> place a capacitor that's larger than parasitic across the choke
> and you'll see that's just a conventional parallel resonant circuit.
That's normal and doesn't prove that we have no transmission line.
"Conventional" circuits are just special cases of transmission lines,
there is no wide and deep gap separating them.
> Experience tells that it is perfectly possible to get a reasonable high
> impedance without
> dealing to reach a resonance.
How high is "high"?
Without approaching parallel resonance (and without a ferrite core)
the best we can get ranges from ~500 Ohm for RG213 to ~2 kOhm
for a thinner conductor.
> Self resonance is hardly a stable condition,
> unless the environment is controlled (closed) and weatherproofed.
Well, that's why we aim to place the parallel resonance
at the geometrical mean of the band in question, in hope
to retain a high enough impedance on both edges.
I've taken the liberty to CC this message to the reflector,
maybe somebody else is also finding amusement in impedance measurements :-)
73,
Sinisa YT1NT, VA3TTN
_______________________________________________
Topband mailing list
Topband@contesting.com
http://lists.contesting.com/mailman/listinfo/topband
|