Amps
[Top] [All Lists]

SV: [AMPS] Linearizing Linears, feedback or feedforward

To: <amps@contesting.com>
Subject: SV: [AMPS] Linearizing Linears, feedback or feedforward
From: Peter_Chadwick@mitel.com (Peter Chadwick)
Date: Tue, 26 Sep 2000 13:47:52 +0100
Tom says;

        >It adds uncontrolled RF feedback with undesired phase shift to the 
        >PA. Feedback level varies with drive power, and from band to band.


Capacitive feedback in grounded cathode circuits can work well - BUT:-

In the 'standard' Bruene neutralising circuit, where a grid tuned circuit has
one end to the grid and the cold end via a capacitor to ground, with the
neutralising cap from plate to the 'hot' side of this capacitor, NFB can be
achieved by suitable choice of capacitor values.

If the phasing effects Tom mentions are not to screw things up, a deal of care
is needed in choosing the capacitor to ground - it MUST have low inductance. The
neutralising gets screwed up if it doesn't too. Both sides of the tuning cap
MUST be floating too.

Downside is that because the value of the neutralising cap goes up, the
effective capacity from plate to ground goes up, which can offer problems at
10metres.

        >The Orr circuit belongs in zero grid current cathode driven tetrodes
>ONLY.

That circuit (or to be precise, a variant of it) can be used with grid current
and grounded grid - the super cathode driven approach uses something like that
although that is with a tetrode. Similarly the semi cathode driven amplifier
-see Pappenfus.

I believe the '220pf to ground caps' business got used first in the 30L1. The
simplified schematic I have suggests that the ratio of grid to filament RF volts
were used for establishing correct tuning and loading, which seems a little
strange. It may be that approach works with 811As, but I agree with Tom that is
isn't a general approach.

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


<Prev in Thread] Current Thread [Next in Thread>