[hpsdr] frequency standard.

michael taylor mctylr at gmail.com
Fri Jun 15 18:11:02 PDT 2007


On 6/15/07, FRANCIS CARCIA <carcia at sbcglobal.net> wrote:
> My point was to make the oscillator remote mounted from the control
> electronics so the user can select the quality of oscillator. I have a
> couple oven controlled VCXOs that will suit my needs for now. I would think

The starting point given by Rick W2GPS and Steve, N7HPR is based upon
Rick's excellent CNS Clock II, which with the suggested top-quality
quartz OCVCXO and quantization ("sawtooth") error correction is not
subject to easily being improved. It also draws from the Reflock II
which is useful for controllable output to met just about any
conceivable need.

Because various VCXO have different parameters, and they may not be
all a single numeric value specified in a datasheet, such as
temperature curves which is dependent upon the physical characteristic
of how the quartz crystal is cut by the manufacturer. For best
performance the oscillator controller has to be "aware" of these
parameters and expected performance of the oscillator it is trying to
steer. I'm not sure if it would be feasible to offer support for user
supplied oscillators.


<http://hpsdr.org/wiki/index.php?title=GIBRALTAR>

<http://www.cnssys.com/cnsclock/CNSClockII.html>

>From the Gibraltar block diagram, the blocks which are digital (or I
think are digital) and could be put in a FPGA are USB I/F, Sawtooth
Correction, Oscillator Steering (fed to a DAC, I think 18 or 20 bits),
and mux  & synchronous divider which has to be designed with close
attention to minimize jitter and remain closely synchronized, or it
will degrade the performance of the outputs.

I haven't tested or measured yet, but I think the only other way
(other than sawtooth correction) to improve the raw output over a GPS
M12M Timing module is to supply the M12M Timing GPS module
Differential GPS RTCM input (it doesn't have SBAS, WAAS support
directly AFAIK), but I am not even sure if that would create a
measurable improvement or not.

-Michael, VE3TIX

 1181956262.0


More information about the Hpsdr mailing list