[Elecraft] More on N6KR method for setting the reference

[Don Wilhelm]

Folks,

I gave a trial run for using Wayne's method for setting the reference.  I
first arbitrarily maladjusted C22 so I could start from scratch.

I found that Wayne's method will get you quickly and easily to a dial
calibration accuracy that is entirely usable for most ham purposes - and
with no equipment other than the K2 itself , a good detector for 'zero beat'
(good well trained 'perfect pitch' ears or Spectrogram, etc.) and WWV or an
alternate standard frequency source.  Repeat runs can produce even better
results - see my ramblings below.

I warmed up the K2 for about an hour, and in the meantime, I plugged a cable
from my soundcard to the speaker jack so I could use Spectrogram to detect
zero beat.  (I had to string up a temporary antenna too since I am running
new coax lines for the shack installation in my new QTH).  Then set about to
do the adjustment.

Set LSB mode and tuned in WWV - used Spectrogram to line up the tones with
my 500 and 600 Hz markers - okay, I know I have WWV properly tuned.

Took the cover off, put in the probe, set K2 to CAL FCTR and tweaked C22
until the BFO and VFO low order digits were the same - easy and fast to do.
I did find out that one should read the BFO first, switch to the VFO and
then adjust C22 - changing C22 moves the VFO frequency more per degree of
C22 adjustment than for the BFO.
I had to go 'round the BFO, VFO, C22 adjustment about 3 times, but it did
not take a lot of time or effort.

Ran CAL PLL and then CAL FIL (I moved the Filter width one notch up and then
down for each filter for LSB, USB, CW and CWr - that is 16 times.  If you
have not been doing that, you have not actually run CAL FIL completely).
This takes more time and effort than setting the reference.

OK, I had a problem - checking after this was completed, I found WWV to be
20 Hz off (tuned at 10000.02) - yes this is within spec, but I wanted to do
a bit better, so I did another trial run.

Since the first time I ended up 20 Hz off, I reasoned that I could alter
things a bit and get it better, so the second time, I set the VFO to 10 Hz
higher than the BFO low order digits.
Repeated CAL PLL and CAL FIL and checked again - ooops, WWV now 30 Hz off
(tunes at 10000.03) - so I went the wrong way with my correction.

I did it a third time, and set the VFO to be 30 Hz lower than the BFO (again
ignoring the high order digit) - my VFO was 14913.62 and the BFO was
4913.65.  Ran CAL PLL and CAL FIL so the K2 would store new numbers in its
lookup table, and rechecked WWV - This time it came right on at 10000.00!!!

I did check 15 MHz WWV and found it to tune 30 Hz low - couldn't hear 20 MHz
WWV, but I found the 5th harmonic of the reference tuned at 19999.93 MHz.  I
attribute the 70 Hz discrepancy to the 10 Hz uncertainty in the K2 steps
when zeroing WWV in the first place, and I could have compensated a bit more
to see if I could get it any closer, but it is good enough for me - my K2 is
NOT a frequency standard even though it has a digital dial, and even if I
have 70 Hz error, the signal fits within my narrowest CW filter, so I can
live with that - just my humble opinion.  Like I have said before, you have
to work at it even harder if you want to make it better.

I have asked Wayne to comment on my findings, but I guess he is resting for
the weekend.  I will post again if I discover anything further from his
comments.

73,
Don W3FPR