[Elecraft] Observation on new Synthesizers for K3

[Al Lorona]

We like to think that a VFO exists only at one frequency -- the frequency on the receiver display.
But in reality every VFO has width, and it occupies not only it's nominal frequency but is "smeared" both lower and higher in frequency, too. This is because of the phase noise of the synthesizer or VFO.
Because of the smearing effect of this phase noise, received signals can also appear wider than they are. The noise floor on either side rises in direct proportion to the synthesizer's phase noise. Dave's screenshots of the P3 spectrograms show this smearing clearly.
An oscillator with less phase noise looks more like that ideal picture we all have in our heads -- of a signal that's infinitely narrow. In the third of Dave's screenshots you can see how the new synths are closer to an ideal oscillator-- the CW signal's width on the spectrogram is much narrower. 
If a signal has lower noise sidebands (whether the sidebands are generated in the transmitter or the receiver... each of them has a synthesizer) then you can enjoy less interference from an adjacent signal. You will also *cause* less interference to your neighbors on the band.
I have no idea of the design of the new synths, but in general to design a synthesizer with low phase noise you have to start with very low noise devices, pay really careful attention to the parts of the phase-locked loop like the Q, feedback, the numeric dithering, the loop filter and various other aspects of the circuit. It's a real art. It appears from Dave's observations that there is a significant and measurable difference.

Al W6LX