...
At these times there is also a little bump in NTPFreqOffset. I can think of two possible causes of this. It could be due to increased interrupt load at these times, causing increased latency in the interrupt function that is called in response to the PPS interrupts. Increased latency in An increase in the delay of response to PPS interrupts should cause NTP to think that the GPS clock has fallen behind the system clock, but the NTPClockOffset at these times is positive, and the slope of NTPFreqOffset is positive, indicating that NTP thinks the GPS clock is ahead.
OrAlternatively, the bump could be caused by increased heating of the system clock oscillator, due to increased CPU load during the file transfers. The sign of NTPClockOffset is consistent with a heating effect, as described above. Also, when the ambient air temperature is very cold, the bump is diminished, or has the opposite slope. So my thinking at this point is that the bump is caused by increased oscillator heating.
...