[Elecraft] BFO Drift

[k6se@juno.com]

When I first got my K2 (S/N 2622) running last May, my biggest beef was
the VFO drift with temperature.  I had grown so used to my FT-1000MP's
stability (whose VFO drifts a maximum of about 10 Hz with extreme
temerature changes with its stock oscillator (which is a non-TCXO) that I
expected the K2 to be as good.

Even in a stable temperature environment, my K2 drifted more than the 100
HZ spec given for the rig by Elecraft and after I griped about it, Gary
Surrency sent me a new pair PLL reference oscillator xtals.  The new
xtals got the rig within Elecraft's specs (barely), but when the shack
temperature changed, the VFO would drift much more than 100 Hz.

From a cold start with the new xtals, 10 MHz WWV will drift from about
9999.05 to 10000.04 MHz on my K2's VFO display (if the shack temperature
is stable).  The 4 MHz oscillator is set to as closely to 4000.000 kHz as
possible, BTW.

I came to accept the VFO's inaccuracy as long as I was careful that I did
not operate near the edge of the band.  The only other ramification is
that the frequencies given on DX packet cluster spots might not agree
with my VFO readout.

However, BFO drift in my K2 is intolerable.  The main reason I bought the
K2 was that I thought it might make an ideal portable rig to use at my
favorite salt lake bed location here in the Mojave Desert during
160-meter CW contests.

The problem with the BFO temperature drift is that, when using narrow CW
filters, the received signal may, or may not, be centered in the filter
passband.  Narrow CW filters are essential for best S/N ratio when
receiving weak signals such as those encountered on 160.

Gary Surrency also sent me a new pair of BFO xtals, but they were even
worse than the originals, so I reinstalled the original xtals.  Even at
home, where the temperature does not change as widely as it does in a
tent trailer at the salt lake bed, signals are outside of the K2's narrow
CW filter's passbands more often than not.

For this reason, I've given up on the K2 and it now gathers dust in a
corner of the shack.

The thread on removing one of the PLL oscillator xtals to alleviate VFO
temperature drift has given me new hope, however.  If it works, this
should cut the BFO drift problem if half, possibly making the K2 usable
for my purposes.

The manual states that the BFO must be able to tune from 4912.7 to 4916.3
kHz, or a range of at least 3.6 kHz.  Because the PLL 12.096 MHz xtals
must have a 10 kHz range (and one xtal works in some cases), this
corresponds to a range of about 4 kHz at 4.915 MHz, so maybe it's
possible.

I would appreciate it if anyone has removed one of the BFO xtals to
reduce BFO drift that they let me know the results.

73, de Earl, K6SE