[time-nuts] Yet another GPSDO project.

[Jim Harman]

Tobias asked,
@Jim Harman
Indeed I know Lars' GPSDO. Can you tell what performance you achieved with
one of those?

It is hard for me to quantify the performance because I don't have anything
better to compare it to. The TIC resolution is 1 nsec.

I can say that it will stay phase locked to the 1 pps from my GPS for
months at a time and if I set the time constant to 1,000 sec, the 300 sec
average error detected by its TIC is within +/- 5 nsec more than 50% of the
time. The maximum error is about 100 nsec.

If I compare the variations in the TIC error from one day to the next, I
can easily see variations of about +/- 20 nsec on a period of just less
than 24 hours caused by the shifting geometry of the satellites overhead. I
am sure this could be improved significantly if I used a a timing-type GPS
instead of the Adafruit one I have.

Lars may have better data in his write-up,

> --
>

--Jim Harman