[Elecraft] OT: crystal ?s...

[John Clifford]

Sorry about the off-topic, but this is about the only place (reflector)
where I expect I can find someone who can answer my questions.  Obligatory
Elecraft reference:  I am using my K2 as a test receiver!

My other rig is a Ten-Tec Omni VI.  It suffers from frequency drift during
QSOs, especially those with high duty cycles such as PSK31 or the other HF
digital modes.  Ten-Tec specs the Omni VI to remain within 50 Hz and it
does... but a 20 or 30 Hz shift down the band during transmit is annoying in
PSK and makes staying in tune a hassle on MFSK.  In the spirit of making my
OTHER rig (Omni VI/Option 1) better (more frequency-stable), I decided to
upgrade the Omni's crystal oven to the later TXCO and also to swap out the
two BFO crystals on the Omni's TX Audio Board for newer ones produced by ICM
that are supposed to be more thermally stable.

The Omni VI uses two BFO crystals (Y1 and Y2).  The LSB crystal (Y1) is
9.000000 Mhz, while Y2 (USB/FSK Mark & Space/CW) is nominally at 9.003000
Mhz... however later Omni VIs use a Y2 at 9.003300 Mhz (300 Hz difference).

The BFO alignment on the Omni VI consists of setting trimmer caps
corresponding to each mode so that the BFO freq is correct according to spec
and puts the transmit and/or receive passband where it is supposed to be in
relationship to the IF.

For LSB, Y1 only needs to be at 9.000000 Mhz (the crystal freq).  For USB,
Y2 needs to be at 9.003000 Mhz, nominally the same as the crystal's
frequency, but it oscillates at a higher freq and must be trimmed down
(adding capacitance via the trimmer) in order to reach the specified freq.
For CW, Y2 needs to be brought down to 9.000400 Mhz (2600 Hz under nominal).

I had to have a total of about 30 pF to get the 9.003000 Mhz USB crystal to
provide a 9.003000 Mhz crystal at the BFO test point.  If I had to add
capacitance, does that mean this crystal really has a resonant frequency
higher than 9.003000 Mhz?

How can you add negative capacitance to a crystal?  Inductance?  The FSK
trimmers have inductors, while the CW trimmer has a resistor... could I
decrease the value of the resistor and thereby lower the base frequency of
the crystal for that mode?

I am wondering how the original crystal, with a higher resonant frequency,
was able to be swung down to a lower BFO frequency with way less
capacitance.  Perhaps my new crystal is not actually a 9.003000 Mhz crystal.

Anyone know of a simple circuit that I can build on a breadboard to
check/verify crystal frequency?

Anyone have any ideas?


(BTW, the unit is much more thermally stable in USB mode and stays with a
couple of hertz during QSOs.  That's more like it!  That is also why I don't
want to put the original 9.003300 Mhz USB crystal back in the rig, as was
suggested by someone at Ten-Tec.)

Thanks,

 - jgc

John Clifford KD7KGX

Heathkit HW-9 WARC/HFT-9/HM-9
Elecraft K2 #1678 /KSB2/KIO2/KBT2/KAT2/KNB2/KAF2/KPA100
Ten-Tec Omni VI/Opt1

email: kd7kgx@arrl.net