Re: [Elecraft] KREF3 External Reference Option: Detailed proposal

2010-09-08 Thread Nige
Wow that was a close one. I was planning to buy a loaded K3 + P3 as a back-up 
radio for my Flex 5K which does have an external reference input. The cart was 
loaded to go but now I know the K3 doesn't have this capability this is very 
disappointing and I'll have to reconsider what to do with my money.

This really is a shame.

Nige.
G7CNF
__
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


Re: [Elecraft] KREF3 External Reference Option: Detailed proposal

2010-09-08 Thread David Pratt
This is coming, Nige; see Wayne's msg on the reflector on 15 June at 
00:15:39 

- - - - - - - - - - - - - 8 - - - - - - - - - - - - - - - - - - -
There appears to be more demand for this than we thought, so we're 
looking into it again.

Note that our planned external reference implementation would not 
require an exact 49.38-MHz reference oscillator signal. It also 
doesn't require a pin-1 control voltage input on the TCXO module itself, 
as suggested earlier. It should work with any existing K3 TCXO, either 
+/- 5 or +/- 1 PPM.

If you're interested, here are the details:

A new REF-lock module would be designed that plugs into the small 8- pin 
connector near the top of the KREF3 PCB. The user's 10-MHz reference 
signal would be routed from the rear panel to the TMP connector on the 
KRE3 (this connector faces forward, into the DSP/ front panel area).

A microcontroller on the REF-lock module would be clocked by the 10- MHz 
reference. If the signal were present, the microcontroller would 
continuously measure (and average) the frequency of the 49.38 MHz 
oscillator, obtaining a value to the nearest 1 Hz. Hysteresis would be 
applied to prevent hunting, so the actual accuracy spec would probably 
be +/- 2 Hz. This translates to about +/- 1 Hz in operating frequency on 
20 m.

Approximately once per second, the module would report the actual 
reference frequency to the K3's main MCU via a slow digital data line 
(labeled 100HZ on the RF and KREF3 schematics). The main MCU would use 
this to calibrate the REF CAL value normally entered by the user. There 
would be no noticeable impact on performance.

If the 10 MHz external signal were not present, the user's normal REF 
CAL value would be used.

Let me know if you have thoughts on this implementation.

- - - - - - - - - - - - - - - - 8 - - - - - - - - - - - - - - - - -

73 de David G4DMP

In a recent message, Nige ni...@hotmail.com wrote ...

Wow that was a close one. I was planning to buy a loaded K3 + P3 as a 
back-up radio for my Flex 5K which does have an external reference 
input. The cart was loaded to go but now I know the K3 doesn't have 
this capability this is very disappointing and I'll have to reconsider 
what to do with my money.

This really is a shame.

Nige.
G7CNF
-- 
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:Elecraft@mailman.qth.net

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


Re: [Elecraft] KREF3 External Reference Option: Detailed proposal

2010-06-17 Thread Trevor Smithers
There appears to be more demand for this than we thought, so we're  
looking into it again.

This is excellent news and I'm sure will be much appreciated, especially with 
the number of 
surplus Thunderbolt GPS and Rubidium 10MHz standards available at the moment.
 
I have no idea of the amount of RD involved in producing this module but at 
which point in the 
development cycle would you expect to be by the end of this year 2010 ?

73
Trevor  G0KTN
__
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


Re: [Elecraft] KREF3 External Reference Option: Detailed proposal

2010-06-15 Thread SM6TUW
Wayne,



Nice workaround for the lack of ”pin-1” .

It seems to offer:



Applicable on all XO:s

Inexpensive

No impact on present XO specification (phase noise etc).

Holdover without EXT REF

Function check by watching REF CAL



Thoughts:



1. The ref oscillator is not actually locked in the above solution.
You have talked of an add-on board for the K144XV which will among other
things contain a PLL locked to KREF3.
There will not be an optimum accuracy between the IF:s unless the K144XV
PLL is designed to lock to the EXT REF instead.
Then there must be a function for holdover in K144XV also.



2. Someone suggested use of a GPS 1pps ext ref. That is useful if the
REF-lock module MCU can perform some averaging over a longer period since
the 1 pps is only accurate to 1ppm (typically).



Questions:


1. What is the minimum PN , spurious and harmonic requirement for the
Elecraft TCXO?

I have been trying a prototype using AD4001, a VC-TCXO and an Atmel
Tiny.

It is locked to an external GPS derived 10 MHz. Probably this solution
has higher PN than the Elecraft TCXO.
I am also investigating if a Slilicon Labs Si 550 is usable in this PLL.


2.  Are you going to publish the K144XV schematic?


73
Jan
__
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


Re: [Elecraft] KREF3 External Reference Option: Detailed proposal

2010-06-14 Thread Wayne Burdick
There appears to be more demand for this than we thought, so we're  
looking into it again.

Note that our planned external reference implementation would not  
require an exact 49.38-MHz reference oscillator signal. It also  
doesn't require a pin-1 control voltage input on the TCXO module  
itself, as suggested earlier. It should work with any existing K3  
TCXO, either +/- 5 or +/- 1 PPM.

If you're interested, here are the details:

A new REF-lock module would be designed that plugs into the small 8- 
pin connector near the top of the KREF3 PCB. The user's 10-MHz  
reference signal would be routed from the rear panel to the TMP  
connector on the KRE3 (this connector faces forward, into the DSP/ 
front panel area).

A microcontroller on the REF-lock module would be clocked by the 10- 
MHz reference. If the signal were present, the microcontroller would  
continuously measure (and average) the frequency of the 49.38 MHz  
oscillator, obtaining a value to the nearest 1 Hz. Hysteresis would be  
applied to prevent hunting, so the actual accuracy spec would  
probably be +/- 2 Hz. This translates to about +/- 1 Hz in operating  
frequency on 20 m.

Approximately once per second, the module would report the actual  
reference frequency to the K3's main MCU via a slow digital data line  
(labeled 100HZ on the RF and KREF3 schematics). The main MCU would  
use this to calibrate the REF CAL value normally entered by the user.  
There would be no noticeable impact on performance.

If the 10 MHz external signal were not present, the user's normal REF  
CAL value would be used.

Let me know if you have thoughts on this implementation.

73,
Wayne
N6KR

__
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


Re: [Elecraft] KREF3 External Reference Option: Detailed proposal

2010-06-14 Thread Eugene Balinski
10 MHz is an industry standard.  This is goof.  However,
you might want to consider a 1PPS GPS-trained input as
well.   

73,
Gene K1NR  

On Mon, 14 Jun 2010 16:15:39 -0700
 Wayne Burdick n...@elecraft.com wrote:
 There appears to be more demand for this than we thought,
 so we're  
 looking into it again.
 
 Note that our planned external reference implementation
 would not  
 require an exact 49.38-MHz reference oscillator
 signal. It also  
 doesn't require a pin-1 control voltage input on the TCXO
 module  
 itself, as suggested earlier. It should work with any
 existing K3  
 TCXO, either +/- 5 or +/- 1 PPM.
 
 If you're interested, here are the details:
 
 A new REF-lock module would be designed that plugs into
 the small 8- 
 pin connector near the top of the KREF3 PCB. The user's
 10-MHz  
 reference signal would be routed from the rear panel to
 the TMP  
 connector on the KRE3 (this connector faces forward, into
 the DSP/ 
 front panel area).
 
 A microcontroller on the REF-lock module would be clocked
 by the 10- 
 MHz reference. If the signal were present, the
 microcontroller would  
 continuously measure (and average) the frequency of the
 49.38 MHz  
 oscillator, obtaining a value to the nearest 1 Hz.
 Hysteresis would be  
 applied to prevent hunting, so the actual accuracy spec
 would  
 probably be +/- 2 Hz. This translates to about +/- 1 Hz
 in operating  
 frequency on 20 m.
 
 Approximately once per second, the module would report
 the actual  
 reference frequency to the K3's main MCU via a slow
 digital data line  
 (labeled 100HZ on the RF and KREF3 schematics). The
 main MCU would  
 use this to calibrate the REF CAL value normally entered
 by the user.  
 There would be no noticeable impact on performance.
 
 If the 10 MHz external signal were not present, the
 user's normal REF  
 CAL value would be used.
 
 Let me know if you have thoughts on this implementation.
 
 73,
 Wayne
 N6KR
 

__
 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

-
Web mail provided by NuNet, Inc. The Premier National provider.
http://www.nni.com/
__
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


Re: [Elecraft] KREF3 External Reference Option: Detailed proposal

2010-06-14 Thread Eugene Balinski
I should not type when tired.  That should have read this
is good, not goof   sorry about that..

73
K1NR

On Mon, 14 Jun 2010 22:04:31 -0400
 Eugene Balinski euge...@nni.com wrote:
 10 MHz is an industry standard.  This is goof.  However,
 you might want to consider a 1PPS GPS-trained input as
 well.   
 
 73,
 Gene K1NR  
 
 On Mon, 14 Jun 2010 16:15:39 -0700
  Wayne Burdick n...@elecraft.com wrote:
  There appears to be more demand for this than we
 thought,
  so we're  
  looking into it again.
  
  Note that our planned external reference implementation
  would not  
  require an exact 49.38-MHz reference oscillator
  signal. It also  
  doesn't require a pin-1 control voltage input on the
 TCXO
  module  
  itself, as suggested earlier. It should work with any
  existing K3  
  TCXO, either +/- 5 or +/- 1 PPM.
  
  If you're interested, here are the details:
  
  A new REF-lock module would be designed that plugs into
  the small 8- 
  pin connector near the top of the KREF3 PCB. The user's
  10-MHz  
  reference signal would be routed from the rear panel to
  the TMP  
  connector on the KRE3 (this connector faces forward,
 into
  the DSP/ 
  front panel area).
  
  A microcontroller on the REF-lock module would be
 clocked
  by the 10- 
  MHz reference. If the signal were present, the
  microcontroller would  
  continuously measure (and average) the frequency of the
  49.38 MHz  
  oscillator, obtaining a value to the nearest 1 Hz.
  Hysteresis would be  
  applied to prevent hunting, so the actual accuracy
 spec
  would  
  probably be +/- 2 Hz. This translates to about +/- 1 Hz
  in operating  
  frequency on 20 m.
  
  Approximately once per second, the module would report
  the actual  
  reference frequency to the K3's main MCU via a slow
  digital data line  
  (labeled 100HZ on the RF and KREF3 schematics). The
  main MCU would  
  use this to calibrate the REF CAL value normally
 entered
  by the user.  
  There would be no noticeable impact on performance.
  
  If the 10 MHz external signal were not present, the
  user's normal REF  
  CAL value would be used.
  
  Let me know if you have thoughts on this
 implementation.
  
  73,
  Wayne
  N6KR
  
 

__
  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
 

-
 Web mail provided by NuNet, Inc. The Premier National
 provider.
 http://www.nni.com/

__
 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

-
Web mail provided by NuNet, Inc. The Premier National provider.
http://www.nni.com/
__
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