[Elecraft] K2 - frequency control

[Geoffrey Mackenzie-Kennedy]

David Woolley (E.L) wrote on Sunday, June 08, 2008 at 10:59 AM

> That would suggest an error in your CAL PLL data.  I'd suggest 
> re-calibrating the 4MHz oscillator then doing CAL PLL.  It may be 
> advisable to recalibrate the filters, as your 4MHz may have been wrong. 
> Normally you only need to move them up one notch and then back down, so 
> that the firmware senses a change and re-reads the frequency.

David,

I believe that your suggestion is a correct explanation for one cause of the 
clicks, but a situation also exists where the clicks disappear in the 
absence of strong signals. In case it might be of interest to the List, here 
follows a comment which I made to Nick this morning.


The clicks are very noticeable with my K2 when tuning between 7160 and 7195
kHz (roughly) if the strength of the BC stations in that part of the band is
much over S9, and they can reach S9 + 70db as measured by the S meter of my
homebrewed receiver, which is calibrated in dbm, also by my Spectrum
Analyser. I have not really checked other parts of 40m and am interested
that you find clicks on 30m as well. However the clicks disappear when the
BC stations become weaker.

>From tests made with a sig gen and Spectrum Analyser I am pretty sure that
the root cause of the problem is the lack of shielding and proper bypassing
in the K2, also pcb layout. It would appear that a strong signal entering
the antenna connector leaks into the VCO and PLL Reference area of the RF
board, and after a process of mixing with one or more of the K2's
oscillators appears as a signal whose fequency is about the same as that of
the PLL Reference. If this rogue signal is strong enough it attempts to take
over control of the VCO loop from the PLL Reference, and the clicks are
caused by these attempts. If the incoming signal is strong enough I have
found that the K2 will lock onto the signal and won't let go until the
receiver has been tuned a Khz or more away, and then lets go with a 'whoop'.
The small jump in frequency that you notice is an example of this 'lock'
behaviour.

73,
Geoff
GM4ESD