Date: Fri, 24 Oct 2003 17:32:33 -0700 From: Szabolcs Marka To: Keith Riles Subject: Re: Timing status? The Pre-S3 Comissioning of LHO Caesium beam clock based timing system is finished. All the atomic clock epics interfaces are up and running at LHO. The sequencer and database code under CVS in project/atomic_clock where: atomic.st -generic sequencer atomic_ex.db - specific database files for ex,ey,mx,my,2klvea,4klvea. The typical way to start the sequencer is: seq &atomic,("ifo=H1,sys=LVEA") The currently used EPICS channels (recording the second trends) are: Alarm Channels: > H2:GPS-LVEA_ATOMIC_FAIL > H1:GPS-LVEA_ATOMIC_FAIL > H2:GPS-MX_ATOMIC_FAIL > H2:GPS-MY_ATOMIC_FAIL > H1:GPS-EX_ATOMIC_FAIL > H1:GPS-EY_ATOMIC_FAIL Measurement channels: > H2:GPS-LVEA_ATOMIC_TIME_DIFF > H1:GPS-LVEA_ATOMIC_TIME_DIFF > H2:GPS-MX_ATOMIC_TIME_DIFF > H2:GPS-MY_ATOMIC_TIME_DIFF > H1:GPS-EX_ATOMIC_TIME_DIFF > H1:GPS-EY_ATOMIC_TIME_DIFF The fiber delay was set to a generic 10uS for the mid stations and 20uS for the ends (~5ns/m). The alarms have been added to the GPS alarm handler, and it is running on the operator station. --- The hardware was installed for the LLO Caesium beam clock based timing system We installed the hardware components of the new atomic clock based timing system at LLO. The mass storage room (MSR) houses the CsIII clock and the Master Clock Assembly of the optical fiber based distribution system. Time is transferred via single-mode fibers to the LVEA and the Mid/End stations, where it is compared to the GPS time. Every component of the MSR assembly is installed, tested and running. The Time Code Translators (TCT), timing comparators and related cabling are installed in the LVEA and the two end stations. The end station installations are tested and running fine. The basic communication to the 162s(EPICS) is not tested as it would require more installations and software changes, as there are slight differences between LLO and LHO. It is not clear whether the changes can be safely finished before S3. Probably not. The lack of communication will prohibit us from recording the timing difference second trends during S3. The LVEA setup indicates a stable ~0.5 us difference between the Cs time and the PEM GPS time. The X end stations shows a stable ~2+20 us difference between the Cs time and the GPS time. It is consistent with the delay we expect due to the fiber delays (~5ns/m). The Y end station measurement unfortunately indicated wild and periodic variations (e-log) http://www.ligo-la.caltech.edu/ilog/pub/ilog.cgi?group=detector&date_to_view=10/15/2003&anchor_to_scroll_to=2003:10:15:12:47:54-smarka Replacing the GPS board fixed the problem. Now the Y end stations shows a stable ~1.2+20 us difference between the Cs time and the GPS time. It is consistent with the delay we expect due to the fiber delays. For now I set the fiber delay for the end stations to the generic 20.00us, to be consistent with the LHO setup. We plan to measure the precise fiber delays after S3, as the distribution system is capable of subtracting it with 5 ns accuracy.