If you look at my earlier message in this thread, you'll see the link to nabble
and my post from 11 years ago where I describe this method. :-)
Wes N7WS
On 6/7/2020 6:03 AM, Randy Farmer wrote:
The best way I've found to calibrate the K3 reference oscillator is to use the
500 Hz and 600 Hz audio tones transmitted by WWV. I put the line out audio
through a sound card and look at it with SpectrumLab. Tune in WWV in USB or
LSB mode and tweak the reference trim until the tones are correct when you
switch sidebands with the dial at precisely XX.000 000 MHz. I've found that
you can't get precise agreement between sidebands, probably because of
quantization limits in the synthesizer, but it will certainly be less than a
couple of Hertz. This is plenty accurate for amateur (or probably any other
kind) of service.
73...
Randy, W8FN
On 6/6/2020 10:20 PM, Bob McGraw K4TAX wrote:
From my take, it is "ham radio" therefore +/-1 Hz. should be good enough for
most operations. After all, the K3S resolution is 1 Hz., +/-1 count as I see
it. I can keep mine +/-2 or 3 Hz on most bands.
I use WWV with the radio in CW mode and CWT on, tune close to WWV and press
SPOT. If it settles on the WWV frequency that's good. If it is off a few
Hz, then I tweak the REF CAL up or down a few Hz until I get the accuracy I
wish by repeating the process several times. Still I find +/- 1 Hz is about
it, even with the high stability TXCO and very adequate warm-up time of about
2 hrs.
If one needs something more accurate, then Don is correct, test equipment is
the way to go. And expect to spend big bucks for good quality equipment that
IS traceable to NIST. If the NIST document or calibration is more than 1
year old, the results will be questionable.
I wrote an article which was published in QST, Sept 2015. It deals with
"Transmit and Receive On Frequency". It shows that digital readouts are just
that, readouts, and they are not frequency determining or measuring
circuits. And when it is accurate on one band it may not have the same
accuracy on another band.
73
Bob, K4TAX
On 6/6/2020 5:20 PM, Don Wilhelm wrote:
With the main K3 fine tuning at 1 Hz steps, I don't know that 0.1Hz or even
0.25Hz doppler shift will matter much in the final result.
My frequency counter is good to 10 exp-9 which equates to +/-0.1 Hz at the
TCXO frequency, so the WWV method provides as good or better accuracy, even
considering the doppler shift possibility.
The main problem is chasing the beat note down to a stable solid note. You
usually can't truly get there, but you can get close enough that you hear
about 10 or 20 seconds between peaks. Close enough for me.
It can be quite expensive to obtain stability better than the K3S in an
analog oscillator. My HP8640B signal generator will do that, but it takes
at least a 3 hour warmup before it becomes stable. Yes, all the internal
enclosures in my '8640 have covers with all the screws installed - that
helps. OK, that is 'old iron', but I am not going to spend several $10,000
for something better. I have better things to do with my money, and no
longer have access to modern lab quality equipment to achieve that kind of
stability.
We have a ham band transceiver - not a precision lab instrument. As long as
we can stay inside our ham bands, that is all that matters to me. I would
not put a carrier exactly on 7,000.00 kHz with any transceiver.
73,
Don W3FPR
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:Elecraft@mailman.qth.net
This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to wes_n...@triconet.org
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:Elecraft@mailman.qth.net
This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to arch...@mail-archive.com
