Sintela DAS Clock Drift Tests 2025

Purpose:
Observe system clock drift on the Sintela DAS system compared to Ubuntu host.

Conducted By: Henthorn

General Setup:
* Sintela system running standalone no network without NTP corrections
* Host laptop connected to the DAS system via Ethernet
* Host laptop has WiFi network connectivity and is NTP corrected
* Periodically record the system time on the two systems simultaneously
* Track the changes in the differences between the systems 
* Save in CSV format

October 2025 Test
* Conducted in the lab at room temperature with sphere open
* Periodically (semi-random) wandered over to the lab a few times a day and made a measurement
* Used ssh combined with shell prompt timestamping to approximate simultaneous sampling
* Went on for about a week
* SintelaDASDrift251005.xlsx
* Summary: 98 ms drift faster in 30.33 hours, or about 0.9 PPM
* That is right in the expected range at room temperature

December 24 - 29 2025 Test
* Conducted in the Bldg G cold room at 6 C then to 4 C with sphere closed
* Used same ssh/prompt timestamping but ran a shell script on the DAS to output the epoch time to nanosecond
* One sample per minute
* Getting the all that data into a CSV file was a pain
* SintelaDASDrift251224.csv
* Observed way more drift at that temperature
* Very linear and predictable
* You can see the change in temperature in the plot
* Between 910 and 940 ms per day faster, or 10.6 PPM
* That is right in the expected range at room temperature (expect 10 - 15 PPM)

January 02 - 05 2026 Test
* Conducted in the Bldg G cold room at 4 C then to room temperature with sphere closed
* Used a Python client/server setup where the DAS sends the system time to host
* One sample per minute
* Host reads DAS time and logs that with it's own system time
* Getting the all that data into a CSV file is very easy
* SintelaDASDrift260102.csv
* Observed way less drift at that temperature
* Very linear and predictable
* You can see the change in temperature in the plot
* 5 to 6 hours to reach steady state inside the sphere
* Back to < 1 PPM (faster)
* That is right in the expected range at room temperature (expect 10 - 15 PPM)