Subject: RE: [AMPS] Bias for SB-220
Date sent: Fri, 4 Jun 1999 03:23:21 -0700
From: Rich Measures <measures@vc.net>
To: "'amps'" <amps@contesting.com>
> >Rich says:
> >
> >> The bottom-line is that there should not be a situation where the bias
> >> can be in the non-linear/Rx mode when the RF relays are in the Tx mode.
> >>
> >
> >I must admit that I've never thought it through very carefully before,
> >but unless the bias is switched before the RF gets there, you must be
> >effectively pulse modulating.
>
> ? indeed.
Bunk. The claims that all kinds of splatter is produced are
nonsense, mostly based on not understanding the effects of IMD.
In the first place, even if the circuit is done incorrectly, it is the high
level IMD with a long duty cycle that creates problems. For an
example of this, just measure two-tone IMD in a class C PA with
adequate loading on the tubes. It's possible to get -20 to -25 dB
third order IMD out of a class C PA at high power levels, because
the envelope distortion at cross-over occurs at such low power
levels. Besides that, the duty cycle of the IMD is short so the
adjacent channel energy is low.
That's why poor designs like the G2DAF system with lots of
crossover distortion sometimes are accepted as "OK". Unless you
are on a quiet band the low level IMD created by poor bias linearity
is washed out in the noise.
Of course I don't think it is a good idea to populate the world with
systems that do that, even if under many or most conditions
ambient noise at the receiver covers the low-level IMD.
Use a PA that clips at nearly full power, where the duty cycle is
long and power levels high, and that is what cause the stuff that
drives others up and down the band wild.
-10 dB IMD at one watt is not near as destructive or disruptive as
-15 dB IMD at one kilowatt. If it was, we all be sending back out
solid state rigs for refunds. Problems depend more on WHERE on
the envelope slope non-linearity occurs than anything else,
because that affects the total power in the spurious products.
You don't have to suffer with any deterioration in IMD at all when
using electronic bias if you understand basic electronics.
If the tubes are biased into a linear region but at very low quiescent
current, and if the threshold of the switch is much less than the
point of non-linearity, and if the response time of the switch is
much faster than the envelope rise time (which is limited by filter
bandwidth), no harm could ever occur.
Only if the builder is shortsighted enough to use a design that
forces tubes into full cutoff between words, and if the response time
or threshold is too low, problems can occur.
The bulk of the clicks Rich reports, if he actually hears them and if
it is not just psychosomatic clicking, more than likely come from
the envelope overshoot causing the PA to flattop on leading edges,
where power levels are high enough that the very short IMD pulses
can have enough energy to overcome noise.
73, Tom W8JI
w8ji@contesting.com
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|