[time-nuts] TAPR TICC timestamp anomalies ("time dilation" parameter)

[Matthias Welwarsky]

Dear all,

I've been spending a lot of time with the TAPR TICC recently and been getting 
to know the instrument better. I have a question for those of you that have 
used or are using the TAPR TICC for any kind of measurements:

My setup typically involves a reference clock source for the TICC, this is 
normally an OCXO, and a 1PPS signal from the DUT, like, a GPSDO. The OCXO and 
the DUT are not locked against each other so naturally there is a phase drift 
between them.

In a current session, the OCXO drifts around 300ps per second against the 
GPSDO, that means every 330 or so seconds (every 100ns of phase shift) the 
"time1" counter wraps and the "clock" counter steps up or down and this 
creates a small discontinuity in the timestamps. The "time dilation" parameter 
in the TICC configuration is supposed to compensate for that effect.

This compensation better be "perfect", because, depending on the degree of 
"misalignment" there will be a very distinct "bump" in the ADEV graph at Tau = 
<time1 wrap interval>. 

The questions for you TICC users are: Have you noticed this effect in your 
measurements? Do you typically change this parameter, or do you leave it at 
the default value? If you do change it, what is your method for finding the 
correct value?

Regards,
Matthias