Frequency reference for the K3 directly off a GPS

[Sverre Holm (LA3ZA)]

Some recent GPS modules (ublox Neo 7) have a programmable output that in addition to the 1 pps output can be programmed even to output 10 MHz. This output has a lot of jitter as it seems to be derived from a 48 MHz clock which is divided down in such a way as to generate a mix of short and long cycles but so that the average frequency is maintained accurately at 10 MHz.

A description can be found here
https://sites.google.com/site/g4zfqradio/u-blox_neo-6-7 and performance is partly described here:
https://www.febo.com/pipermail/time-nuts/2014-August/086257.html

My question is regarding the K3 external reference input. Is this source good enough to drive the K3 reference? Can the K3 clean up the jitter internally, or does the K3 require a source with much less phase noise?

[Matt Zilmer-3]

Hi Sverre,

This type of 10 MHz output has short term stability problems and isn't
suitable for direct drive of a K3 (imho).  It *could* be used, if
integrated over a long time constant (thinking many seconds) to drive
a D.O., but that would take additional circuitry.  It might be quite a
project.  

You're probably better off with A GPSDO or a Rubidium source.  Even a
plain OCXO would work if you can keep it calibrated.

73,
matt
W6NIA

On Thu, 13 Aug 2015 06:36:31 -0700 (MST), you wrote:

>Some recent GPS modules (ublox Neo 7) have a programmable output that in
>addition to the 1 pps output can be programmed even to output 10 MHz. This
>output has a lot of jitter as it seems to be derived from a 48 MHz clock
>which is divided down in such a way as to generate a mix of short and long
>cycles but so that the average frequency is maintained accurately at 10 MHz.
>
>A description can be found here
>https://sites.google.com/site/g4zfqradio/u-blox_neo-6-7 and performance is
>partly described here:
>https://www.febo.com/pipermail/time-nuts/2014-August/086257.html
>
>My question is regarding the K3 external reference input. Is this source
>good enough to drive the K3 reference? Can the K3 clean up the jitter
>internally, or does the K3 require a source with much less phase noise?
>
>
>
>
>-----
>Sverre, LA3ZA
>
>K2 #2198, K3 #3391,
>LA3ZA Blog: http://la3za.blogspot.com,
>LA3ZA Unofficial Guide to K2 modifications: http://la3za.blogspot.com/p/la3za-unofficial-guide-to-elecraft-k2.html

Matt Zilmer, W6NIA
--
"Always store beer in a dark place."  -R. Heinlein

______________________________________________________________
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]

[Jim Brown-10]

Take a look at what KA6SQG did to sync both audio and RF from GPS for
this five-site 440 MHz simulcast repeater system.

http://www.wb6ece.org/technical-info.html    Play the youtube video
while flipping through the slides. You'll need headphones to get
acceptable audio from the youtube.

I live in the area covered by this system, and can attest that it works
quite well.

73, Jim K9YC

On Thu,8/13/2015 6:59 AM, Matt Zilmer wrote:

> Hi Sverre,
>
> This type of 10 MHz output has short term stability problems and isn't
> suitable for direct drive of a K3 (imho).  It *could* be used, if
> integrated over a long time constant (thinking many seconds) to drive
> a D.O., but that would take additional circuitry.  It might be quite a
> project.
>
> You're probably better off with A GPSDO or a Rubidium source.  Even a
> plain OCXO would work if you can keep it calibrated.
>
> 73,
> matt
> W6NIA
>
> On Thu, 13 Aug 2015 06:36:31 -0700 (MST), you wrote:
>
>> Some recent GPS modules (ublox Neo 7) have a programmable output that in
>> addition to the 1 pps output can be programmed even to output 10 MHz. This
>> output has a lot of jitter as it seems to be derived from a 48 MHz clock
>> which is divided down in such a way as to generate a mix of short and long
>> cycles but so that the average frequency is maintained accurately at 10 MHz.
>>
>> A description can be found here
>> https://sites.google.com/site/g4zfqradio/u-blox_neo-6-7 and performance is
>> partly described here:
>> https://www.febo.com/pipermail/time-nuts/2014-August/086257.html
>>
>> My question is regarding the K3 external reference input. Is this source
>> good enough to drive the K3 reference? Can the K3 clean up the jitter
>> internally, or does the K3 require a source with much less phase noise?
>>
>>
>>
>>
>>

______________________________________________________________
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]

[Sverre Holm (LA3ZA)]

I am aware that a GPSDO etc is fine. But is it overkill for the K3?

What I am trying to determine is the minimum requirement for the oscillator that will drive the K3's external reference input. The documentation says that the K3's phase stability is not affected by the external reference. That, too me, sounds as if there is some internal averaging or whatever in place already.

Matt Zilmer-3 wrote

You're probably better off with A GPSDO or a Rubidium source.  Even a
plain OCXO would work if you can keep it calibrated.

[Sverre Holm (LA3ZA)]

I have done some more experimenting now with my poor man's 10 MHz reference. The ublox Neo-7 GPS is a lousy frequency reference in most respects, except that it is accurate in keeping exactly 10 million periods per second. My experience is that the K3 has enough internal averaging to deal with that.

Using a ublox Neo-7 (USD 12 on Ebay) and a 74HC04 as a driver to feed the 10 MHz to the 50 ohm input of the K3EXREF the K3 accepts the input and I see the star in REF*CAL blinking.

The reference frequency now starts at 49,379,680 and stabilizes to 49,279,650 after some 15-20 minutes. My initial setting for the reference was 49,379,640 which I determined manually when the K3 was new in 2009 or so.

I am going to write this up on my blog with schematics eventually.

Sverre Holm (LA3ZA) wrote

Some recent GPS modules (ublox Neo 7) have a programmable output that in addition to the 1 pps output can be programmed even to output 10 MHz. This output has a lot of jitter as it seems to be derived from a 48 MHz clock which is divided down in such a way as to generate a mix of short and long cycles but so that the average frequency is maintained accurately at 10 MHz.

A description can be found here
https://sites.google.com/site/g4zfqradio/u-blox_neo-6-7 and performance is partly described here:
https://www.febo.com/pipermail/time-nuts/2014-August/086257.html

My question is regarding the K3 external reference input. Is this source good enough to drive the K3 reference? Can the K3 clean up the jitter internally, or does the K3 require a source with much less phase noise?

[Sverre Holm (LA3ZA)]

I've got some questions about the type of GPS module I'm using. It is a ublox NEO-7M. The nice feature of this unit is that it has a programmable output which nominally excites an LED at 1 pulse per second, but which also can be programmed to 10 MHz. That output is not connected, so one needs to make a connection to it oneself.

Sverre Holm (LA3ZA) wrote

I have done some more experimenting now with my poor man's 10 MHz reference. The ublox Neo-7 GPS is a lousy frequency reference in most respects, except that it is accurate in keeping exactly 10 million periods per second. My experience is that the K3 has enough internal averaging to deal with that.