Expt. 201814 BASIN – Issue Report

	•	Nodes were deployed with the Handheld Terminals (HHTs) on 2018-07-20 UTC between 20:44 to 23:39. The scheduled start time was 2018-07-21 11:30 UTC at which time the nodes performed startup “segment” tests, acquired GPS and began acquisition. Starting at approximately 14:00-14:30 UTC on 2018-07-21 the nodes were then transported to field locations ~50 miles from the initial deployment location and installed at individual field sites. 

	•	When the nodes were moved to the field as above, the next time a GPS lock was acquired with strong enough satellite signal the nodes performed a reset because of the location being different and causing an issue with the node’s timing solution. A reset creates a gap in data acquisition, and a discontinuity in drift correction.
	The overall consequences are that there were 1 or 2 gaps in data acquisition for all but 11 of the 57 nodes returned. And, also only 11 nodes out of 57, that had valid GPS timing from sometime on 7/22 through sometime on 8/25, although it was a different 11. 
  Two spreadsheets are accompanying this issue report. One spreadsheet lists, for each node, the start and end time for each segment of acquired data. And the other spreadsheet lists, again for each node, the start and end time of valid GPS timing during each segment of data acquisition. Valid timing was considered to be intervals where the drift did not exceed 1/2 the sample rate. Within those intervals though, frequently nodes had numerous Acq Log entries with invalid timing solutions. 
  In addition to having been moved after acquisition was started, many of the nodes were installed in the field in such a way that their GPS did not have good satellite reception. Either the nodes were buried too deeply, or there were buildings or vegetation obstructing the sky view, or perhaps both. 
  The combination of the nodes having been moved, along with poor satellite reception, meant that in some cases nodes did not have any GPS locks until removed from the field, at which time a reset occurred because the location was different and thus there were no drift corrections for the entire time they were in the field. In other cases, the nodes reset at the time of field installation, and then whether it had valid GPS timing over the course of the deployment or not depended only on the satellite reception. 

	•	When nodes start acquisition they perform a number of self-tests. These can serve as verification that the nodes were level and properly functioning when deployed. When acquisition is started at somewhere other than the installed location, that verification is absent. And, in instances where a node failed the self-tests at the initial deployment location, whether the node continued in a poor state or whether it was installed in a way that it could function properly is unknown. The quality of the data is called into question. Units that had failed segment tests are listed below.
	•	S/N 1884 - 2nd and 3rd segments failed Phone Natural Frequency on all three channels ranging from 3.1 to 4.1 Hz, minimum passing value is 4.5Hz, and Phone Impedance on Ch 2 at 2848 ohms, max passing value is 2820 ohms.
	•	S/N 1893 – 1st segment failed Phone Natural Frequency on all three channels at 5.66-5.68 Hz, maximum passing value is 5.5 Hz.
	•	S/N 1899 – 1st (only) segment failed Phone Natural Frequency on all three channels at 5.80 -5.96 Hz, maximum passing value is  5.5 Hz.
	•	S/N 1928 – 1st segment failed Phone Natural Frequecy on all three channels at 2.59-3.87 Hz, minimum passing value is 4.5Hz, and Phone Impedance on Ch 1 at 4000 Ohm, maximum passing value is 2820 Ohm.