On Nov 21, 8:24*pm, "NoSPAM" wrote:
"exray" wrote in message

...

*In terms of old transmitters from the 20s/30s...In a crystal oscillator I
understand the concept of setting the oscillator output tank to favor the
harmonic from the crystal. *(Stop me if I'm wrong already...)

*But in a doubling amplifier stage am I counting on having enough harmonic
content at the input or am I creating the harmonic with the non-linearity
of the amplifier?

*TIA
*-Bill *WX4A

Hi Bill.

Remember that single ended frequency multiplier stages are usually operated
in Class-C where the nonlinear operation of the stage produces the
harmonics. *In a Class-C stage, the grid (of the tube since we are talking
about vintage transmitters) is biased such that the plate current only flows
in short pulses. *The narrower the pulse width, the greater the harmonic
generation of the stage. *If you look in the old RCA Transmitting Tube
Manual, there is a design procedure where the "conduction angle" of the tube
is chosen for proper harmonic generation.

Frequency doubling is unique in that two Class-B stages may be used in a
push-push arrangement. *Here the grids are driven in push-pull while the
plates are connected together in parallel. *The resulting waveform will
essentially be the equivalent of full-wave rectification of the input
signal. *Without going into Fourier series, the resultant waveform only
contains even harmonics of the input signal while the fundamental driving
frequency is cancelled out.

Fortunately I was already a ham operator when my high school math class
taught Fourier series *. *I immediately saw the practical value of this
mathematical concept and it made good sense to me. *Your question is a good
one and reading some of the tutorials on Fourier series (do a Google search)
will be very useful to your understanding of harmonic generation and
intermodulation distortion. *I hope that my simple explanation will start
you in your own exploration.

* * 73, *Barry L. Ornitz * WA4VZQ

* More years ago that I care to admit! *:-)

Wow, there's a name I haven't seen for a while. I must be frequenting
the wrong groups. I was just thinking about you a couple days ago.
Hi Barry!

More about what Barry wrote: in the limit as the conduction angle
goes to zero and you generate a very narrow pulse of current, the
harmonics end up all the same amplitude. That's for an impulse of
zero width. Unfortunately, given limited amplitude of the current in
that very narrow pulse, the total energy becomes small. As you widen
the pulse, you'll see that the "comb" of harmonics no longer has
constant amplitude, but the amplitudes as you go up the "comb" (higher
in frequency)drop, and there will be a frequency at which they go to
zero, and then increase again (and go to zero again, and increase
again). The magnitudes follow a sin(x)/x shape, for perfectly
rectangular pulses. This becomes interesting for a the design of
frequency multiplier stages: if for example you want to get x4 out of
a stage you better NOT run it at a conduction angle that results in
nulling of the fourth harmonic! I think I was bit by this a time or
two in my youth when I didn't understand this. (I'm working on
something right now where I want that comb of harmonics to be all very
nearly equal amplitude up to about 100MHz, and that tells me how
narrow the pulse must be.)

Cheers,
Tom