Content-Transfer-Encoding: 8bit
Message-ID: <65AECDF1F89AD411900400508BFC869F0D746C@pdw-mail-1.dera.gov.uk>
From: "Talbot Andrew" <ACTALBOT@dera.gov.uk>
To: rsgb_lf_group@blacksheep.org
Subject: LF: RE: Re:amplifiers
Date: Fri, 8 Dec 2000 08:41:02 -0000
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1; format=flowed
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

The parasitic diode is no good in this configuration for catching the
spikes from what I call the transient soak choke, I use separate clamp
diodes up to the rails effectively 'crosing over' the centre tapped
choke - see the circuit in the new LF Experimenters Book.

I once met this impedance inverting effect the hard way with an earlier
PA and it shows up very clearly in simulations.  Here I rely on the
series tuned tank circuit to perform most of the filtering; that has no
Z inverting action but with the expense of slightly extra loss.   By my
reckoning no extra filtering is then necessary as second harmonic is
practically non existant due to the square wave drive,  and third
harmonic is 26dB down to start with due to the tank and then seriously
reduced further by the antenna loading coil.  I don't think the Decca
transmitters used any filtering other than the tank circuit.  

I also have it on good authority from a transmitter designer, that the
classic PI filter with input C is VERY bad news for using at the output
of  MOSFET PAs.  The shunt cap presents a low impedance to harmonics and
increases circulating currents to stress the devices unnecessarily.
Better to use an input inductor in a Tee (or even order) filter even if
it does mean having to use an extra inductor.  A recent QEX magazine
(the ARRL amateur experimeters journal) also had an excellent article on
this and other aspects of PA design, last year.  The author here went
even further and used diplexing filters on the output of an HF PA - in
my opinion a bit OTT.

Andy  'JNT

> 
> Yes, the "totem-pole" configuration is sort of self-clamping 
> if MOSFETs
> are used. Did you add external (low Vf) diodes or do you rely on the
> intrinsic diodes in the FETs?
> 
> >As far as I can see, the only potential device destroyer is a low
> >impedance load or short circuit.
> <snip...>
> >By the look of the network, A detuned antenna will reflect back as an
> >increased load impedance giving an inherent fail safe mechanism.
> 
> The lowpass filter between the amp' and the antenna can turn 
> things upside
> down when a severe mismatch occurs. For example, a simple pi 
> filter with
> a Q of 1 (C1 across, L along, C2 across, all reactances equal 
> to line Z)
> will "invert" the impedance like a lambda/4 length of 
> transmission line.
> In case the filter output is open (antenna way off 
> resonance), L and C2 forms
> a series resonant circuit which makes the filter input look 
> like a dead short at
> the design frequency. On the other hand, A short on the 
> filter output causes the
> input to look like an open circuit (C1 and L in parallel resonance)...
> The impedance inverting properties of a simple pi filter can 
> be useful when the
> antenna is fed via a tap or link on the loading coil but for 
> series fed antennas
> I'd choose two cascaded pi filters (like the one on G3YXM web site).
> 
> 73
> Johan SM6LKM
> 
> 
> 
> 


-- 
The Information contained in this E-Mail and any subsequent correspondence
is private and is intended solely for the intended recipient(s).
For those other than the recipient any disclosure, copying, distribution, 
or any action taken or omitted to be taken in reliance on such information is
prohibited and may be unlawful.