Amps
[Top] [All Lists]

Re: [Amps] Some thoughts on a transistorized legal limitamplifier...

To: amps@contesting.com
Subject: Re: [Amps] Some thoughts on a transistorized legal limitamplifier...
From: Radio WC6W <wc6w_amps@yahoo.com>
Date: Sun, 14 Oct 2007 13:40:23 -0700 (PDT)
List-post: <mailto:amps@contesting.com>
Combined replies to comments:

>Marv - Why not use Class-E for the final amplifier if
you are going to use 
>EER? Efficiency 90% . You would only need to
dissipate 150 watts in the heatsink vs 600?
>73 - Todd WD4NGG

Hi Todd,
   Wellllll, to my limited understanding, Class E
provides poor utilization of (expensive!) FET Drain
voltage capacity, requiring something like 4X supply
voltage on that specification plus, given the
restrictions on the associated L's & C's, it may be
difficult to QSY over all the required bands.   

   My personal preference is for Class D with a 3rd
harmonic trap but, that would run up the cost quite a
bit as at this power level a vacuum cap is preferred
for said trap.

******************************

>I think you'd need a feedback system - polar or
cartesian loop, 
>for example. The AM/PM in the output stage will be
severe, and 
>probably beyond simple pre-correction in open loop.

>You can buy the whole feedback system integrated on a
single chip 
>these days, but all the uses I'm aware of are VHF and
higher so I 
>don't know if it will work at HF.

>Steve

Hi Steve,
   The chips you mention which work at VHF are
employed with amplifiers exhibiting nanosecond group
delays and as part of a complete transmitter.

   Back to HF, the DSP provides limiting to create the
phase signal which is sent through the exciter at a
constant (30 watt) level.  This will avoid the AM to
PM problem.

   This is basically the Kahn EER scheme, from 1952,
kluged to run with a typical transceiver in the signal
chain.  If it worked back then, it should still work
now! 

   Just guessing but, considering all filters
involved, the group delay on a typical transceiver is
probably in the 100's of microseconds, and certainly
not constant over the transmitters bandwidth.  The
other function of the DSP is to delay equalize the
amplitude path for the envelope restoration.  The
magnitude of this delay would preclude closing the
loop with any kind of feedback.  

   Adaptive equalization might prove beneficial, if
the "natural" open loop response isn't good enough for
rock & roll and amateur radio.

   It is technically easier, though it does require
quite a few more parts, to fashion a complete DSP
based exciter, with constant group delay & other
improved specifications, to drive the power stage
directly but, that's not germane to the original
question... not that anyone has ever wandered off
topic on this reflector.  :-)

73 & Good afternoon,
   Marv WC6W

http://wc6w.50webs.com/






*


       
____________________________________________________________________________________
Building a website is a piece of cake. Yahoo! Small Business gives you all the 
tools to get online.
http://smallbusiness.yahoo.com/webhosting 
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps

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