[Elecraft] K2 S/N 4787 First QSO

[Ron D'Eau Claire]

Fraser G4BJM wrote:

Sorry, I don't understand this point.  The K2's I've tested fall down on LO 
phase noise performance.  Also note that the K2 measurements on the Sherwood

web site state that the receiver performance is limited by phase noise.  I 
agree the RX works very well, but I've attracted comments on the air (when 
running QRO) re transmitted phase noise.  Surely the good receive 
performance is primarily due to having good selectivity near the front end.

Phase noise seems to be a taboo subject on here...  Is there a way to bring
it in line with other top rigs?

-------------------

You're the first that I've heard to make such comments, Fraser. 

I have not tested the K2 myself. I was referring to the design approach the
K2 uses that allows for lower phase noise than a fully-synthesized
oscillator, but at the cost of somewhat less absolute frequency stability
and dial accuracy. 

As you know, reducing the frequency at which phase correction is applied to
a PLL reduces the phase-generated sidebands or "phase noise". If you have a
fully-synthesized oscillator must stay within 1 Hz of the reference
oscillator, it has to be corrected often enough that it will never drift
more than 1 Hz or it'll start "walking" across the band in 1 Hz steps. That
involves very frequent corrections and resulting noise. 

Elecraft uses a PLL that tunes in only 5 KHz steps. It does not need to
correct nearly as often as a fully synthesized oscillator, hence less phase
noise. 

Of course, a local oscillator that tunes in 5 kHz steps isn't very useful on
the Ham bands so, instead of using a very precisely controlled reference
oscillator, the K2 uses a variable frequency oscillator for the reference
oscillator. More accurately, it uses a variable frequency *crystal*
oscillator, or VXO, for the reference oscillator. The VXO is tuned by a
varicap over a 5 kHz range to fill in the gaps in the PLL tuning. The result
is a very quiet PLL with continuous frequency coverage.

The down-side of that approach is that because the reference oscillator is
designed to change frequency, it will tend to drift. A fully-synthesized PLL
can use exotic crystal oscillators in temperature-controlled ovens to
provide a much higher degree of stability than any variable-frequency
oscillator can achieve. So, in terms of absolute frequency stability, the K2
is not as stable as some fully-synthesized rigs. Still, in my experience,
the K2 needs offer no apologies for frequency stability, even in very
demanding situations such as PSK31. 

If you are experiencing phase noise problems in either transmit or receive,
I'm sure Elecraft would like to hear about it. I'm not personally familiar
with the Sherwood web site.

I've never seen a subject relating to Elecraft equipment that wasn't eagerly
discussed and dissected here on the reflector, although comments do get
missed from time to time. Certainly, no such queries are "taboo". Indeed, a
great many of the design improvements to the Elecraft rigs have come
directly from initial criticisms published right here, followed by a lot of
discussions about ways to improve the rigs that eventually led to a design
change. Moreover, every time such a change was made, an upgrade kit was also
made available to all current owners so their rigs could keep pace with the
performance of the very latest new units being built.   

Ron AC7AC