At 11:32 AM 2000-02-14 -0700, measures wrote:
>Preventing grid-bias from becoming more negative increases grid current.
As the proper grid bias value shifts more negative with grid current, the
plate current rate of change with drive power becomes less linear; this can
be easily seen using the tube characteristics curves. The issue becomes
more important for speech amplification when the time constant of the power
supplies cannot follow speech due to high source impedance. For example,
assume a high speech peak which causes a peak positive-going grid current
that causes the grid bias to shift more negative, such as happens when
using a series-pass grid bias regulator. Then the grid current peak goes
away but unless the time constant of the bias supply is sufficiently-fast
as to allow the bias to immediately bleed back to the correct value, it
will cause the tube to be further cutoff when the next speech element
occurs. This, in addition to Escr supply variations, is often the cause of
"buckshot".
> Grid current means distortion.
Yes, for class A and AB1 at a desired output power level, the amount of IMD
(harmonic distortion) will increase beyond that which is desired (and
expected). However, the tube merely shifts into class AB2, then B, and in
an extreme case, C. Distortion increases but the amount of additional
distortion will depend upon the tube characteristics. The 4CX1000-series
was designed for excellent IMD characteristics in zero-grid-current class
AB1 while others, such as the 4CX800A, are designed to generate acceptable
levels of IMD when a small amount of grid current is drawn. In addition, of
course, there are V/UHF amplifiers where grid circuit inefficiencies are
high enough that *some* grid current is almost always drawn. The levels of
IMD generated as a result are an unfortunate byproduct but are often
acceptable, particularly when the exciter itself is already generating
significant levels of IMD to which the contributions of the amplifier
become essentially negligible.
>Why not adjust the grid bias potential
>so that there is zero grid current with full drive? ZSAC is set by adj.
>screen potential.
It is never a good thing to allow tube element voltages to change with the
drive power; one loses control over the operating characteristics of the
tube. In CW mode, this might be an acceptable solution, perhaps, since
there is no modulating waveform (other than the bandwidth caused by the
rise and falltimes of the keyed driving signal). But with a time-varying
speech waveform, the resulting IMD caused by peak element currents cannot
be predicted. (Actually, in CW mode and at normal keying speeds, there is
likely to be some key click generation immediately surrounding the carrier
which level may or may not prove to be unacceptable depending upon a number
of factors.) G3SEK and others have shown that even when the modulation
waveform is a two-tone test signal, that peaks of the two signals combining
simultaneously can cause unexpected-high IMD levels due to peak tube
element currents which cause momentary voltage sag of power supplies with
high source impedance. The end result will be a propensity toward
generating buckshot on voice peaks even though on average, the IMD level
may well be acceptable.
73, Steve K0XP
--
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
|