RT130

Leap Second's effect on EPIC RT-130 and Q330

Summary:

Q330:

Data time starting at 2017:001:00:00:00 is +1 second from truth for 1-2 minutes (seen in controlled tests) until a clock correction, followed by a gap of 2-40 seconds per channel depending on sample rate.

RT-130:

Data time starting at 2017:001:00:00:00 is +1 second from truth for 15min - 9:42hrs (seen in controlled tests) until a clock correction. Some data (~6secs in tests) around the time correction is mistimed, sometimes 11 years in the future, or missing.

Important update to rt2ms program for processing RT130 data

This is to inform you about an important software update needed for processing of Reftek RT130 data in EPIC experiments. We recently discovered an issue with running rt2ms on the second card installed in the RT130.  Sometimes insufficient header values are being written by the RT130 and then rt2ms passes over these packets potentially resulting in a gap on one or more channels (typically less than 1 hour on higher sample rate data, larger gaps on 1sps data).  The EPIC Software Group has patched rt2ms to handle the missing packet headers and write all of the data that are on the card.  We highly recommend that you download and install the updated rt2ms program if you are processing any RT130 data.

Jan. 5, 2015

EPIC recently received reports of RT130’s failing when using two RefTek 4G CF cards at moderately low temperatures during an Antarctic experiment. We have duplicated the failure in our lab and in our testing the failure was associated with a specific version of RefTek well board (RT526D). Although we have not had failures in our testing with earlier or later versions of the board, our sample set is small and we cannot rule out problems with these boards when using two 4G CF cards. We have communicated with RefTek about the issue and at this time they are in agreement with these findings.

This note is to alert all past and current users of RefTek RT130 hardware that these instruments have been experiencing GPS failures in-field. To date, the failures are geographically confined to East Africa and symptoms include intermittent or total loss of timing.

It is worth noting that in reviewing close to two years of repair records, 1303 GPS have been returned to PIC from 69 experiments, excluding those in East Africa, and the total number of GPS units that needed any type of maintenance or repair was 30. There is no evidence at this time that there is a global problem with our RefTek GPS.

We are recommending that all users of EPIC RefTek RT130 dataloggers review their data and log files for signs of GPS failure that are consistent with the failures seen in East Africa. To identify failure behavior please visit this page for a brief tutorial. If you identify failures in your data, please contact passcal [at] passcal [dot] nmt [dot] edu and include "RT130 GPS failure" in the subject line and unit serial numbers in the body of the email.

In order to accept your passive-source data for archiving please comply with the procedures and standards below. EPIC staff are available to assist you with meeting the requirements of archiving your SEED data as stated in the EPIC Data Delivery Policy. 

Regarding the archiving of SEED data, the policy may be summarized in three points:

ETS, or 'episodic tremor and slip', is a recently discovered phenomenon in seismic research. Similar to earthquakes but much smaller in magnitude, ETS events are associated with the subduction zone of some convergent plate boundaries. One such region, the Cascadia subduction zone under the Puget Sound, is the focus of an experiment by Ken Creager's group at the University of Washington (UW), called Array of Arrays.

DSP CLOCK DIFFERENCE -1 SECS and 0 MSECS messages mark the beginning of timing errors in the RT130 data, due to the SiRF clock’s erroneous time shift. The SiRF goes back and forth between correct and incorrect time over periods of hours or days. The RT130, and hence your data, will follow suit, generally with a lag of 4 hours.

‘POSSIBLE DISCREPANCY’ messages will be produced whenever the SiRF and the RT130 are not in sync. Four consecutive ‘POSSIBLE DISCREPANCY’ messages will reset (JERK) the RT130 time.

If the clock is being power cycled, the clock is locking and the RT130 is phase locking with the SiRF each time it is powered, the 4-hour lag is produced.

Stations Affected: RT130 connected to a 130-GPS/01 with SiRF ublox Trimble module installed (aka SiRF-type)

Description of Problem: Starting on July 31st 2008 (julian day 213 2008) the SiRF-type started to shift in time by a second.  Since the 31st July 2008 the mysterious time shifts (back and forth) continue. At present SiRF has no explanation for these time shifts.

Identifying SiRF GPS: