[Elecraft] K2 PLL Optimization

[Lyle Johnson]

Eddi,

> >    ok, maybe I'll get an answer here. So far I completely fail to under-
> > stand how a varactor in an oscillator's frequency determinant circuit
> > can work without introducing a load of "dirt" effects.
> >
> >    A varactor's capacity is controlled by the (reverse) bias, the
> > higher the voltage, the lower the capacity, right? The rf voltage
> > across the tuned circuit's capacitor (said varactor) constantly adds
> > or subtracts itself to/from the bias voltage once per half wave. To
> > me this means a varactor controlled oscillator must output a whole
> > frequency band, i.e. lots of noise sidebands.


This is part of the circuit design.  You want a fairly large control voltage
(DC bias) and a fairly low amplitude RF voltage at the point the varactor is
applied.  This helps reduce distortion.

If you look at the K2, you'll see the VCO uses series connected varactors.
The DC bias voltage is the same at D23 and D25, for example, but they are in
series-opposing to the AC voltage in the tank circuit.  Thus, as one diode
sees its voltage decrease due to RF, the other sees its voltage increase.
While the diodes aren't linear, to the extent they are this cancels the
effects of the AC voltage on the overall capacitance, thus reducing
distortion.

You'll also note there are two diodes in parallel in each leg of the
circuit, for a total of four diodes.  Why is this?

Ah, that is an exercise left to the reader.  Just keep in mind the goals of
the circuit designer for this part of the radio...

73,

Lyle KK7P