[Elecraft] K2/100 VFO cal problem - need help

[Lyle Johnson]

Hello Tim!

> ... How
> can I get the variation on the higher frequencies to be closer to the Icom
> and how come the variation is not the same on all bands?

The Icom probablyuses a synthesizer that is slaved to a single master
oscillator (this is an assumption on my part based on other Icom rigs).  It
may or may not be on frequency and may vary with time, temperature, etc.
Unless you have it slaved to a known frequency standard, it may be slightly
off in frequency.

(In my shack, I have a rubidium oscillator and a pair of GPS-slaved HP
frequency standards.  These all match within a fraction of a Hz at 10 MHz,
and I use one of the HP's as the reference for my frequency counter, signal
generators, etc.)

The K2 uses a simpler scheme.  The upside is a very clean, low-phase-noise
oscillator system; the downside is slight frequency error and drift with
temperature, time, etc.

When you perform Cal PLL, if the MCU reference oscillator is *exactly* 4.000
MHz, then you will get the best accuracy of PLL calibration.

However, the reference oscillator is simply a crystal oscillator that is
"pulled" by driving a varicap from a DAC by the MCU.  The varicap has
temperature variations, and the DAC is dependent on the voltage regulator(s)
in the K2.  It is not intended to be a frequency standard, only a reasonably
accurate method of setting frequency.  The specs on the K2 are +/- 30 Hz
over a 500 kHz range when calibrated, and drift of <100 Hz from cold start
at +25C.

As to the error changing with bands, this is because the VCO runs at a
different frequency range on 15m (21-5= 16 MHz) than it does on 40m (7+5= 12
MHz), so any PLL reference error is also multiplied.

I suggest you read the theory of operation section of the K2 manual, study
the schematic and block diagram as you read -- then read it again :-)  The
answers are there...

73,

Lyle KK7P