Mike, you know what it's like, when not understood, re-write it to suit
yourself! Am I guilty of this?
I had always taken "linear" to mean that the amp can accept any
modulation, and amplify it without distortion. Thus a "linear" amp can
amplify FM, CW, AM or SSB without re-biasing. The particular gain, or
quest for constant gain when doing so, is relevant only to the extent
that is should be worthwhile!
A "non-linear" amp probably operates in class C, E or similar. It is
"non-linear" because it cannot amplify certain signals without
distortion. It doesn't matter if the gain changes with changing input,
provided the amp is not distorting, which is the original thread of this
discussion (I think!).
Now, what's all this about constant gain?
G0MJW wrote:
>
> > A linear does not demand constant gain at any power level, within the
> > context of this discussion
>
> You have an interesting definition of linear then. An "amplifier for
SSB"
> may not demand constant gain, but by definition, a linear amplifier
does. It
> also requires linear phase but I seldom see a phase/amplitude AM/PM
> conversion or group delay plot for an amateur PA.
Group delay is normally of concern where the modulating signal has wide
bandwidth as in TV or data. Group delay when threading a couple of kHz
of audio through a ham voice band amp is going to be low, and in any
case lost in the general roar (read mic input..).
I'm not defending, in any way, the DAF biasing scheme, but as someone
suggested on this thread, the initial howls may have been directed at
the moon.
> The DAF circuit relies on the envelope of the drive signal. If it can
peak
> hold well enough then it will appear to be linear. Hence the good 2
tone
> test results. It will also work well for speech as long as the speech
has
> sufficient energy to maintain the screen potential nearly constant. The
IMDs
> will be a function of the modulation envelope and the time constant of
the
> screen supply.
>
> A better way is to use a detector that only biases the valve on when
there
> is a signal to amplify, but to ensure no nasty switch on transients, it
> really needs a delay line.
Does the delay line have a soft turn-on as well?
> This is easy with DSP and is something that
> modern rigs could usefully add to their feature list. The DSP could
also do
> useful speech filtering, pre-distortion and compression.
My ICOM 746 already lays claim to DSP audio tailoring on transmit, but
my 6m neighbours would probably dispute this fact. And the mic gain is
right down, consistent with achieving PEP against no, or minimum, ALC
indication.
Actually, when 6m opens up, we all generate a rich porridge of IMD, and
devil take the hindmost. It seems to be an effective way of elbowing
some space, judging by the way its used.
Dunno about the DSP-on-compression.
> Mike
Ciao.
Ian ZS6BTE
--
FAQ on WWW: http://www.contesting.com/ampsfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|