In article , budgie
writes:

On Fri, 12 Mar 2004 15:32:23 +0000, Ian Bell wrote:

Paul Burridge wrote:

Hi all,

Is there some black magic required to get higher order harmonics out
of an oscillator?
I'm only trying to get 17.2Mhz out of a 3.44Mhz source and am thus far
failing spectacularly. I've tried everything I can think of so far to
no avail. All I can get apart from the fundamental is a strong third
harmonic on 10.32Mhz, regardless of what I tune for.

In RF circles, the 'normal' way to do this would be a simple Class C
amplifier with a collector load tuned to the fifth harmonic. In calls C,
conduction only occurs for a small fraction of a cycle which produces a
correspondingly higher proportion of higher harmonics than a square wave.

I've been waiting for someone to post this. I would only add "The drive
level,
and the bais point, will vary the amount of fifth (or whichever) you will
see."

It's as common as noses in RF, as Ian pointed out. Just look at the average
two-way radio prior to frequency synthesisers. Crystal freqs were multiplied
this way in transmitter chains and for receive injection, although use of
fifth
wasn't especially common because you normally had enough design control to
use the more efficient *2, *3 or *4.

Fifty years ago that was mostly true and multiplier stages rarely went
beyond the 4th harmonic. Two notable exceptions, though -

A circa 1950 design by General Electric for an 1800 MHz radio relay
terminal used a SEPTUPLER via a 2C39 planar triode. [7 x multiplier]
Roughly 260 MHz input following a buffer stage from a 5th overtone
crystal oscillator. Used in both transmitter and receiver panels ("dish-
pan" style, chassis = rack panel) with the receiver septupler driving
the mixer. Fussy to tune but stayed there once tuned.

Another G.E. design of the early 1950s used Locked Oscillators in a
TV broadcast local color subcarrier supply. Locked oscillators operate
on integral multiples of the input and their use was almost extinct back
then. That was deemed necessary in the G.E. design for an 11 x
multiplier, the highest direct multiple I've encountered. A locked
oscillator can also operate as a divider as G.E. did. The 3.58 MHz
crystal could also be phase-locked to a network feed color burst.

However, all those multiplier types went the way of the dinosaur when
PLLs operating directly at the desired frequency came into being.
There isn't any advantage to using those old "exotic" technologies
other than in restoration for nostalgia's sake.

Quintuplers CAN be made, but, so far, Paul hasn't explained
enough specifics about his circuit, or how he is sensing any 5th
harmonic for any of us to get a good handle on a possible aid.

Note: Lacking any spectrum analyzer, a wide-range HF receiver
with an S meter can be an indicator...but such needs to be checked
against a calibrated signal generator for compensation of varying
S meter indication versus input levels. That's what I use for
checking HF levels (Icom R70) and it has been calibrated against
a reasonably-known-level RF source.

Len Anderson
retired (from regular hours) electronic engineer person