Frequency Accuracy and stability

[Bob McGraw]

I ordered my K3S with "KTCX03-1-F TXCO 1ppm f/w correction to 0.5ppm".  
K3S-F s/n 10163

I use CW mode and CWT mode and tune to each of the WWV frequencies.  By
pressing SPOT the radio resolves to the WWV frequency but will show any
error either + or - of the WWV frequency.   By tuning up about 25 Hz and
then pressing SPOT and then tuning down about 25 Hz and using SPOT one
can easily see the error amount.    If all is correct, it will return to
the WWV frequency +/- 0.0 difference.

I check 4 or 5 WWV frequencies and note any discrepancy.  I then can
adjust the REF CAL up or down as needed and then re-check the WWV
frequencies as described above.

I find this method is preferable to the "zero beat" method and assures
the radio on all bands is +/- 2 Hz or better.   Good enough for me.

73

Bob, K4TAX


______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

[Elecraft mailing list]

I count beats.  One beat in 1 second and I'm off 1 Hz.  One beat in two seconds and I'm off 1/2 Hz.  Easiest done if the side tone and received signal are adjusted to be close in amplitude and there's little/no QSB.   Zero beats, and the frequencies would be exactly the same, but the practical limitation is probably 1/4Hz in good conditions.  The only things required are a reference, like WWV, and my ear.

That is, after all, what the term zero beating means, adjust frequencies to be the same until the beats stop.

Lou W7HV



    On Wednesday, May 19, 2021, 3:30:49 PM MDT, Bob McGraw <[hidden email]> wrote:  
 
 I ordered my K3S with "KTCX03-1-F TXCO 1ppm f/w correction to 0.5ppm".  
K3S-F s/n 10163

I use CW mode and CWT mode and tune to each of the WWV frequencies.  By
pressing SPOT the radio resolves to the WWV frequency but will show any
error either + or - of the WWV frequency.   By tuning up about 25 Hz and
then pressing SPOT and then tuning down about 25 Hz and using SPOT one
can easily see the error amount.    If all is correct, it will return to
the WWV frequency +/- 0.0 difference.

I check 4 or 5 WWV frequencies and note any discrepancy.  I then can
adjust the REF CAL up or down as needed and then re-check the WWV
frequencies as described above.

I find this method is preferable to the "zero beat" method and assures
the radio on all bands is +/- 2 Hz or better.   Good enough for me.

73

Bob, K4TAX


______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]  
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

[Don Wilhelm-5]

Lou,

I find that viewing the output on an audio spectrum analyzer (I use
Spectrogram) helps, one can see the beat to a lower frequency that
normal hearing.  Just watch the amplitude of the signal - when the beat
goes to as close to zero as you desire, you are "there".  I have found
that I can easily get as close as 1 beat in 5 seconds, which is 1/5 Hz
or .2 Hz in the final calibration.
There is a limit to the DAC steps that may keep you from getting to an
absolute zero beat - this is not an analog thing, but then it is more
stable than an analog VFO, so there is some tradeoff.

73,
Don W3FPR

On 5/19/2021 6:54 PM, Louandzip via Elecraft wrote:
> I count beats.  One beat in 1 second and I'm off 1 Hz.  One beat in two seconds and I'm off 1/2 Hz.  Easiest done if the side tone and received signal are adjusted to be close in amplitude and there's little/no QSB.   Zero beats, and the frequencies would be exactly the same, but the practical limitation is probably 1/4Hz in good conditions.  The only things required are a reference, like WWV, and my ear.
>
> That is, after all, what the term zero beating means, adjust frequencies to be the same until the beats stop.
>

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

[Dave Hachadorian-2]

I tune to 10,000,002 and adjust REF CAL for 20 beats in 10 seconds, or as
close to 20 as I can get, with the granularity of the ref cal steps.  In the
last ARRL FMT, all of my measurements were within 1 Hz, using only a K3
without even a TCXO.

Dave Hachadorian, K6LL
Yuma, AZ

-----Original Message-----
From: Louandzip via Elecraft
Sent: Wednesday, May 19, 2021 3:54 PM
To: Elecraft Reflector
Subject: Re: [Elecraft] Frequency Accuracy and stability

I count beats.  One beat in 1 second and I'm off 1 Hz.  One beat in two
seconds and I'm off 1/2 Hz.  Easiest done if the side tone and received
signal are adjusted to be close in amplitude and there's little/no QSB.
Zero beats, and the frequencies would be exactly the same, but the practical
limitation is probably 1/4Hz in good conditions.  The only things required
are a reference, like WWV, and my ear.

That is, after all, what the term zero beating means, adjust frequencies to
be the same until the beats stop.

Lou W7HV



    On Wednesday, May 19, 2021, 3:30:49 PM MDT, Bob McGraw
<[hidden email]> wrote:

I ordered my K3S with "KTCX03-1-F TXCO 1ppm f/w correction to 0.5ppm".
K3S-F s/n 10163

I use CW mode and CWT mode and tune to each of the WWV frequencies.  By
pressing SPOT the radio resolves to the WWV frequency but will show any
error either + or - of the WWV frequency.   By tuning up about 25 Hz and
then pressing SPOT and then tuning down about 25 Hz and using SPOT one
can easily see the error amount.    If all is correct, it will return to
the WWV frequency +/- 0.0 difference.

I check 4 or 5 WWV frequencies and note any discrepancy.  I then can
adjust the REF CAL up or down as needed and then re-check the WWV
frequencies as described above.

I find this method is preferable to the "zero beat" method and assures
the radio on all bands is +/- 2 Hz or better.   Good enough for me.

73

Bob, K4TAX


______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

[ANDY DURBIN]

"I tune to 10,000,002 and adjust REF CAL for 20 beats in 10 seconds, or as close to 20 as I can get, with the granularity of the ref cal steps."

Zero beat has only one solution.  Doesn't your method have two solutions, only one of which is correct?

73,
Andy, k3wyc
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

[Dave Hachadorian-2]

Good point.  One needs to do a careful zero beat cal first, to make sure you
are in the ballpark.  You can verify later by tuning to 9,999,998 and
counting the beats in 10 seconds. The number of beats on the two frequencies
should add up to 40.

Incidentally, by counting the number of beats in 10 seconds on 10,000,002
(or 9,999,998), you can estimate your residual frequency error in tenths of
a Hz (beyond the capability of the dial display).  For example, if there are
23 beats in ten seconds, your radio is telling you that you are 2.3 Hz from
zero beat, so you are actually tuned to 10,000,002.3 Hz.  That info can be
used to correct your results in the Frequency Measuring Test (FMT), but is
of no other practical use that I can think of.  It was fun to beat some of
the big boys in the last FMT, using only a stock K3 without even a TCXO!

Dave Hachadorian, K6LL
Yuma, AZ

-----Original Message-----
From: Andy Durbin
Sent: Thursday, May 20, 2021 4:57 AM
To: [hidden email]
Subject: [Elecraft] Frequency Accuracy and stability

"I tune to 10,000,002 and adjust REF CAL for 20 beats in 10 seconds, or as
close to 20 as I can get, with the granularity of the ref cal steps."

Zero beat has only one solution.  Doesn't your method have two solutions,
only one of which is correct?

73,
Andy, k3wyc
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]