[time-nuts] GPS tick compared to WWVB and WWV ?
paulswedb at gmail.com
Wed Apr 9 21:26:29 EDT 2014
Hello to the group. Still working the wwvb d-psk-r.
A long time ago I did an early experiment looking for the phase flip on
wwvb. I used a GPS tick but at the time the wwvb spec was not in its final
form and I never looked for the flip 100 ms from the tick. Went 25-50 ms.
Kind of call it a gps assisted d-psk-r.
Well I took a look at it again and its been an interesting experiment.
But the first question I have do all GPS receivers essentially tick at the
same time within the 1 ms range.
I assume they do for both navigation and time transfer.
My ublox receiver is 8.6 ms in advance of the wwvb and wwv signal and even
further in advance that at what I suspect is the propagation delay through
the receiver. The fluke207 is 13 ms and the HP 3586 is 14 ms late.(The HP
has far more processing delay than the Fluke).
If the GPS is true than to look for the phase flip you must add propagation
delay + receiver delay + 100 ms wwvb delay to flip or about 118 ms from the
GPS tick. Make sense?
The other thing that really jumps out at me is that using the falling edge
of the wwvb 60 Khz is really a very poor approach to finding the flip. As I
have now measured the falling edge depending on signal levels can be 50 ms
one way or another.
So much for the simple cmax chips as a useful tool directly applied to the
phase flip of wwvb.
It can be still useful as a source of time.
More information about the time-nuts