K3: REF IN

[Julian, G4ILO]

I recently used the frequency measurement facility of the new WSPR software to calibrate my K3, and got into a discussion with an amateur who turns out to be to be the former Head of Standards in British Telecom. He pointed out a few things that are probably obvious to the more technically minded of you already, like the fact that even if I get the calibration spot on at that moment in time, even the 0.5ppm specification of the TCXO option means that the calibration could vary by up to 25Hz at 50MHz.

I live 10 miles from the 60KHz MSF frequency and time standard transmitter so it would be rather cool to build a receiver for this and lock the K3 reference oscillator to it. I see that there is a blanked off hole for an external frequency reference on the back of the K3. What would be necessary to make use of this?

Is the local oscillator for the 144MHz transverter module derived from the main K3 master oscillator, so it would also benefit from the ability to lock the reference oscillator?

[David Pratt]

Page 17 of the K3 Owner's Manual (Rev D2) shows an SMA connector for
"Input for external standard frequency reference" for which a KREF3-EXT
option purports to be available eventually.  I understand from what has
been said earlier is that this will be for a 10MHz standard so you will
need to use the 60kHz MSF signal to lock a 10MHz TCXO or use a GPS,
Rubidium or Caesium standard.

Elecraft are busy preparing the K144XV for despatch and developing the
P3, so we might have to wait a while for the KREF3-EXT despite the
option being shown in the early editions of the K3 manual.

I am told that patience is a virtue.

73

David G4DMP

In a recent message, "Julian, G4ILO" <[hidden email]> wrote ...
>
>I live 10 miles from the 60KHz MSF frequency and time standard transmitter
>so it would be rather cool to build a receiver for this and lock the K3
>reference oscillator to it. I see that there is a blanked off hole for an
>external frequency reference on the back of the K3. What would be necessary
>to make use of this?
--
David G4DMP
Leeds, England, UK
------



______________________________________________________________
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

[gm3sek]

David Pratt wrote:
>Page 17 of the K3 Owner's Manual (Rev D2) shows an SMA connector for
>"Input for external standard frequency reference" for which a KREF3-EXT
>option purports to be available eventually.  I understand from what has
>been said earlier is that this will be for a 10MHz standard so you will
>need to use the 60kHz MSF signal to lock a 10MHz TCXO or use a GPS,
>Rubidium or Caesium standard.
>

The schematic of the KSYN3 board shows a header (J4) with connections to
the 10MHz input, the 49.380MHz output from the existing oscillator
module and a V_ctrl line going back to the oscillator module. This
suggests that the KREF3-EXT module will have a phase-locked loop to lock
the oscillator to 10MHz.

(Also on J4 is a mysterious "100Hz" bus which wanders around many other
areas of the K3... anyone know what that is?)

However, phase locking to an external reference still doesn't mean that
the K3 will be precisely on frequency.  Due to the internal arithmetic
of the K3's synthesis and mixing scheme, the frequency you see on the
display is not quite the frequency you actually get. As Wayne pointed
out on 17 November, "this translates to a VCO step size of about 0.5 Hz
on 160 m, 1 Hz on 20 m, and 2 Hz on 6 m. Other bands are between these
values."

What the KREF3-EXT module should be able to deliver is complete freedom
from frequency drift, for almost all practical purposes. There could
also be great communication value in being able to lock a pair of K3s to
the same global GPS standard, even when they are half a world apart

In the meantime, the K3 already has the most important kind of frequency
and phase locking - between its own two receivers for true diversity
reception.


--

73 from Ian GM3SEK
______________________________________________________________
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

[Jan Holmer SM6TUW]

Since some time I am working on a solution just for fun ( http://n2.nabble.com/KREF-EXT-Option-tc2461086.html#a2461086 ).
My GPS-derived clock was built from VE2ZAZ design and it works fine ( it is also used for other purpose in lab).  PLL is a ADF4001 programmed from a AVR ATtiny2313.
So far the system is running stand alone with a VCXO external to the K3 since my KTCXO3 does not have a voltage control connection. It is certainly possible to modify the KTCXO3 to a VCTCXO but I didn´t dare to do that yet. Maybe Elecrafts supplier can provide such a beast?

If you are planning to use the K144XV as IF for higher order transverters it would be important to get this reference accuracy.

73  Jan

Ian White GM3SEK wrote

David Pratt wrote:
>Page 17 of the K3 Owner's Manual (Rev D2) shows an SMA connector for
>"Input for external standard frequency reference" for which a KREF3-EXT
>option purports to be available eventually.  I understand from what has
>been said earlier is that this will be for a 10MHz standard so you will
>need to use the 60kHz MSF signal to lock a 10MHz TCXO or use a GPS,
>Rubidium or Caesium standard.
>

The schematic of the KSYN3 board shows a header (J4) with connections to
the 10MHz input, the 49.380MHz output from the existing oscillator
module and a V_ctrl line going back to the oscillator module. This
suggests that the KREF3-EXT module will have a phase-locked loop to lock
the oscillator to 10MHz.

(Also on J4 is a mysterious "100Hz" bus which wanders around many other
areas of the K3... anyone know what that is?)

However, phase locking to an external reference still doesn't mean that
the K3 will be precisely on frequency.  Due to the internal arithmetic
of the K3's synthesis and mixing scheme, the frequency you see on the
display is not quite the frequency you actually get. As Wayne pointed
out on 17 November, "this translates to a VCO step size of about 0.5 Hz
on 160 m, 1 Hz on 20 m, and 2 Hz on 6 m. Other bands are between these
values."

What the KREF3-EXT module should be able to deliver is complete freedom
from frequency drift, for almost all practical purposes. There could
also be great communication value in being able to lock a pair of K3s to
the same global GPS standard, even when they are half a world apart

In the meantime, the K3 already has the most important kind of frequency
and phase locking - between its own two receivers for true diversity
reception.


--

73 from Ian GM3SEK
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[e-mail]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

[AD6XY]

Ian White GM3SEK wrote

What the KREF3-EXT module should be able to deliver is complete freedom
from frequency drift, for almost all practical purposes. There could
also be great communication value in being able to lock a pair of K3s to
the same global GPS standard, even when they are half a world apart

And as long as the world is not moving relative to itself the TX and RX could be dead on frequency for low bandwidth modes. Usually things are moving. We could measure the Doppler on the HF signals due to the movement of the ionosphere and do many other things...

The KX144 can not be locked to the reference, it uses a different LO. If it were based on the reference oscillator we would not be hearing of crystal supply problems.

It ought to be possible to lock the supplied TCXO - or a replacement to GPS using the 1PPS or 10kHz signals from a GPS receiver. That might be more useful to most amateurs than a 10MHz reference input.

As an interim solution - I wonder if Wayne could produce a routine to automatically correct the K3 based on a user settable reference frequency? It could be WWV or a GPS locked beacon, a shack based 10MHz source. If that is done after the K3 has warmed up it should stay within a Hz or so for hours.

Currently I do this manually for the transverter bands - where it matters for me. I tune to my 70MHz beacon in CW mode and trim the XV offset to zero beat with the CW tone. It requires iteration to do this. Measuring an offset in frequency is the sort of thing the DSP could do much better and quicker than I can.

[Julian, G4ILO]

AD6XY wrote

Measuring an offset in frequency is the sort of thing the DSP could do much better and quicker than I can.

That's a pretty neat idea. That's what the WSPR 2 calibration routine does - measure an audio frequency to 2 decimal places. The K3 DSP clock is probably more accurate than a computer sound card.

[Jan Holmer SM6TUW]

 There will be a choice according to Wayne:

...."We have made provisions for this on an optional daughter board that
plugs into the main K144XV module. This option board will have at least
three features:
(1) sub receiver AUX RF patch (to let the sub RX
receive on 2 m even if the main RX is on a different band);
(2) 144 MHz
low-level I/O (for higher-band transverters);
 (3) a PLL to lock the
K144XV's crystal oscillator to the K3's 49.38 MHz reference oscillator.

This option will be available somewhat later than the K144XV itself,
and the price has not yet been set. It will be very easy to install".....

I am looking forward to have more info about that option!

73 Jan

AD6XY wrote

The KX144 can not be locked to the reference, it uses a different LO. If it were based on the reference oscillator we would not be hearing of crystal supply problems.

[Jan Holmer SM6TUW]

Any progress??

Jan

[Brett Howard]

This is very interesting!  I'd not heard about this board but I did
see all the extra connection holes in the K144XV shield!  I'm very
interested to get the feature to be able to use the subRX on 2M while
the main is not!  This is one of the things I was hoping to be able to
do with the unit.  Should allow me to wait for a sked or even monitor
a repeater if need be while scanning around on other HF bands...

~Brett (N7MG)

On Sat, Dec 5, 2009 at 2:55 AM, Jan Holmer SM6TUW <[hidden email]> wrote:

>
>  There will be a choice according to Wayne:
>
> ...."We have made provisions for this on an optional daughter board that
> plugs into the main K144XV module. This option board will have at least
> three features:
> (1) sub receiver AUX RF patch (to let the sub RX
> receive on 2 m even if the main RX is on a different band);
> (2) 144 MHz
> low-level I/O (for higher-band transverters);
>  (3) a PLL to lock the
> K144XV's crystal oscillator to the K3's 49.38 MHz reference oscillator.
>
> This option will be available somewhat later than the K144XV itself,
> and the price has not yet been set. It will be very easy to install".....
>
> I am looking forward to have more info about that option!
>
> 73 Jan
>
>
>
> AD6XY wrote:
>>
>>
>> The KX144 can not be locked to the reference, it uses a different LO. If
>> it were based on the reference oscillator we would not be hearing of
>> crystal supply problems.
>>
>>
>>
>
> --
> View this message in context: http://n2.nabble.com/K3-REF-IN-tp4105110p4117002.html
> Sent from the [K3] mailing list archive at Nabble.com.
> ______________________________________________________________
> 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
>

______________________________________________________________
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