[time-nuts] FE-5680A Mechanical Question

Charles P. Steinmetz charles_steinmetz at lavabit.com
Fri Jan 13 15:36:05 EST 2012

Bert wrote:

>we would be talking +- one step and using different rate but reading the
>frequency over 1000 seconds would be my answer.

Rather than dithering the existing 7e-13 steps, perhaps it would be 
simpler to adjust the 5680A step size?  Presumably, the C-field is 
being controlled by the current or voltage output of a DAC, and the 
7e-13 steps represent the LSB of the DAC.  From the reports so far on 
the list, the DAC output range is much larger than necessary (I think 
someone said ~ +/- 2 Hz ??).  If you attenuate the DAC output by 10x 
in the analog domain, you would have a +/- 0.2 Hz range with 7e-14 
steps (or, if you like, attenuate by 7 for 1e-13 steps, or by 14 for 
5e-14 steps).  You would have to be a little careful about the tempco 
of the implementation (particularly, you may need to add some 
low-tempco bias to get the range centered on 10.000000000 MHz again), 
but I wouldn't think it would be so critical as to be beyond the 
ability of the home time nut.

Best regards,


More information about the time-nuts mailing list