Reports until 14:48, Tuesday 27 November 2012
H1 CDS
david.barker@LIGO.ORG - posted 14:48, Tuesday 27 November 2012 (4773)
CDS Maintenance

WP# 3576,3577,3578,3581

busy maintance day today, four separate work permits!

WP3576 IOP SUS Watchdog bypass time extension. Following extending MC2's bypass time on h1iopsush34, today we did the same for h1sush2a (MC1 and MC2). This is an extension of the original WP. I burt restored MC1,3 to the hourly autoburt at 6am. Bypass times are now 21,600 (6 hours).

WP3577 DAQ Memory expansion. In h1dc0, h1nds0 and h1nds1 we replaced 6x2GB (12GB) DIMMS with 3x8GB(24GB). Went very smoothly.

WP3578 found that re-arranging the full height PCIE cards did indeed fix the problem of OAF,LSC,ASC causing dolphin crashes.

old new
Dolphin RFMX
RFMY RFMY
RFMX Dolphin
OneStop OneStop

Basically Dolphin and RFMX cards were swapped. This new PCIE arrangement means that power up of the computer does not crash all Dolphin'ed front ends, and RFM Card0=XArm Card1=YArm. h1lsc0's spectracom half height PCIE card was replaced, which fixed its timing problem. Second bad IRIG-B card so far.

WP3581, following the dismantling of the Newtonian Noise array at EY, we powered down h1pemauxey and h1tcsey and removed their channels from the CDS/DAQ systems. These were H2OAT systems and will not be ran for aLIGO.

h1susauxh34. We did a bit more investigation with this system. Its IOChassis was power-less, but DC power cord and front switch were GOOD and ON. Power cycle of front rocker switch powered the IOChassis back, but reboot of h1susauxh34 caused the IOChassis to de-power again. Interestingly the time from CPU restart to IOChassis power loss varied, and seems to be related to the cool down time of the IOChassis. We will investigate further.

h1isiitmy was now added to the DAQ following the mem upgrade, no problems.

h1susitmy was attempted to be added to the DAQ, but we exceeded a 120,000 channel limit (num fast chans + num slow chans + num testpoints). For now we are only storing the fast sus itmy channels and none of the slow sus itmy channels to keep below the limit (currently running with 117,137). LLO has increased this limit to 250,000 we are working on doing the same.

Jeff Kissel is building the burt safe.snap and filter files for SUS ITMY, based on ETMY.