> ---------- Initial message -----------
>
> From : owner-amps@contesting.com
> To : amps@contesting.com, "'Maurizio Panicara'" <i4jmy@iol.it>
> Cc :
> Date : Fri, 29 Sep 2000 08:22:37 +0100
> Subject : RE: [AMPS] Feedback path 3-500Z
>
>
> Mauri says:
>
> >how, and where, they are connected
>
> 1. 50 uH in parallel with the combination of plate tank tuning
capacitor and
> series coupling cap and series bypass cap.
Going from PS to plate:
Several uF HV cap filter is in parallel with 1000 pf Bypass, reactance
is several uF//1000pf, a like a short (-j0) on VLF.
The 8 uH inductor is in series with the previous found load, 8 uH is
about j10 Ohm at 200 Khz.
But +j10 Ohm is in parallel with the second 1000 pF (-j796 Ohm @ 200
Khz), and equivalent impedance is about j9.7 Ohm.
The 50 uH inductor is in series, Xl is +j62.8 Ohm at 200 Khz, and in
series with +j10 Ohm the equivalent is +j72.4 Ohm.
Until now, on plate is applyed a +j72.4 Ohm reactive load, value that?s
already a full safety against feedbeck for VLF oscillations unless an
opposite -j72.4 reactance would resonate it.
Of course there is a blocking capacitor in series from plate to PI and
its reactance at 200 KHz is around -j800 Ohm.
Then, what is after (in series) with blocking cap must be +j872.6 to
resonate plate at 200 KHz.
In other words, better to say in numbers, we need that PI and antenna
load looks like a 694.7 uH equivalent inductor to resonate plate at 200
KHz.
Neglecting the reactance of 1-20 uH PI coil at 200 Khz (because
changes with band), but having plate and load capacitors that are in
parallel, what as to be in parallel is an inductance (+J) that has to
be bigger than 694.7 (exact value depend by settings and switched band)
Assuming no antenna is connected, the HV safety choke parallel to the
antenna output could be the only guilty device able to rise inductive
reactance of this lag up to such high value.
It?s worth to point out that HV safety chokes are generally placed
after the PI network where impedance is already down to nominal 50 Ohm
and there is no reason for having them too big. First because
unnecessary, second because they should be able to manage some power to
be effective if blocking cap fails, third because higher the inductance
higher the chance the self resonance is within the HF range.
When an antenna is connected things are theorically more complex
because HF antenna impedance can be transformed to high reactive values
if one has long enough lines (in term of WL and fractions like quarter
waves).
Actually I don?t see very raccomandable to use lines whose lenght
ranges in the area of 375m (1230 FT), a quarterwave at 200 KHz.
> 2.The plate tank capacity is the capacity of the tuning capacitor
plus the
> shunting effect (via the tank coil) of the loading capacitor shunted
by the load
> reactance (NOT 50 ohm resistive, but the complex load presented by an
antenna at
> the low frequency)
Agreed that an HF antenna is not 50 +j0 at VLF, the feed point
impedance of an HF antenna at VLF is much less than one ohm and
largely -j reactive (i.e an open dipole) or few Ohm and slightly
inductive +j in case of closed loops, antennas with baluns, matching
devices like hairpin, etc. etc.
As previously explained, unless the used line is quite long and acts
like an impedance transformer there is no chance an HF antenna will
become so higly reactive (+j) to resonate plate load.
73,
Mauri I4JMY
> It is a complex network, with a number of points of resonance
(resonance being
> defined as V and I applied are in phase)
>
> Whether or not the impedances are such as to produce sufficient gain
and total
> circuit phase shift to allow oscillation is another, rather more
indeterminate
> problem: I make no claims in that respect.
>
> 73
>
> Peter G3RZP
--
FAQ on WWW: http://www.contesting.com/FAQ/amps
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
|