Topband: Transceiver Key Clicks

[Paul Christensen, Esq.]

> Does anyone have first hand knowledge of the K2/100 keying characteristics
> and whether it has a click problem? I am also interested in the Ten-Tec
Omni
> VI+ keying and whether it has a click issue, or not?

The stock Omni VI+ has a very sharp and discontinuous leading edge.  The
trailing edge is fine.  Interestingly, the Omni V had much better waveform
shaping but in the analysis I conducted back in 1997 and 1998, the Omni V
achieved its superior symmetrical key shaping at the expense of ALC power
spikes when running at a power level less than 100-watts (more on this
below).

Ten-Tec responded with the Omni VI and VI+.  In both VI designs, the ALC
circuit was modified slightly.  Ten-Tec placed a heavier emphasis on
minimizing power overshoot at powers less than 100-watts than in keeping the
CW waveform smooth and symmetrical.  Also, early Omni VI's allowed the
operator to control the keyed waveform through a small potentiometer located
on the Control Board.  The control never had an effect on the leading
waveform edge and by the time the Omni VI+ was introduced, the majority of
them had fixed resistors in place of the pot.  (i.e., why adjust a pot that
won't properly adjust?)

It can be very difficult to design a transmitter with a symmetrical key
shape while controlling power spikes with an ALC circuit.   The behavior of
ALC tends to react to the onset of power by using feedback.  Only, if the
feedback loop is not fast enough to react to the initial power spike before
the time the ALC has settled, the damage is done.  Alternatively, by making
the ALC respond faster, it tends destroys the integrity of the keyed
envelope.

I am not saying it is impossible to design fast ALC and preserve the keyed
envelope, but the exercise can be very challenging.  It took me close to a
month to find an ALC combination for the Omni VI+ that accomplished this
goal.  It actually took dual ALC time constants; one fast, the other slow in
order to properly manage the ALC.  To make matters worse, all affected
components on the Omni VI+ are SMD and real pain to work with.

Another point worth mentioning is that the transition of the keyed envelope
is much more important in reducing keyed bandwidth than the actual rise
time.  For example, it is possible to construct a 2 ms envelope without any
sharp discontinuities along the keyed edges and maintain perfectly
acceptable spectral bandwidth.  Alternatively, a 10 ms rise/fall with sharp
transitions can still cause keys clicks.  The roughest transitions are
usually found on the leading edge (the ALC dilemma) and appear at the
beginning of the rise of the envelope and again at the end of the rise of
the envelope.  The trailing edge can still be problematic, but is much
easier to deal with than the leading edge.

Another important test presently missing from the QST Product Reviews is to
measure the keyed waveform at full power AND something less than full
power...say 25-50% where many rigs operate into linear amplifiers.  By
viewing the waveform at lower power, it will flesh-out which rigs have ALC
power spikes...a phenomenon that is much more apparent at less than full
power.  The reason we do not see much evidence of a power spike at a radio's
full power is that there's just not much headroom available at this level in
order for a power spike to occur.

-Paul, W9AC