Reports until 02:42, Monday 30 September 2019
H1 SEI (SEI)
corey.gray@LIGO.ORG - posted 02:42, Monday 30 September 2019 - last comment - 08:58, Monday 30 September 2019(52195)
H1 Lockloss: Received Alert for Noisy BRSx

Before a lockloss (9:29utc/ 2:29amPDT), noticed "BRSx is noisy" on Diag Main.  

Then H1 broke lock.

Followed by EX ISI trip & TMSx trip.

BRSx is waivering between DAMPING  & "BRS Bad" for the Damping Control Monitor.  Will investigate more after I try to restore EX ISI & TMS.

Comments related to this report
corey.gray@LIGO.ORG - 02:52, Monday 30 September 2019 (52196)

9:51 Took OBSERVATORY MODE to CORRECTIVE MAINTENANCE since this is an out-of-the-norm lockloss & BRSx looks to be in a BAD state.

corey.gray@LIGO.ORG - 04:18, Monday 30 September 2019 (52197)

Follow-up on BRSx:  

Not sure why BRSx is in a BAD state, but back in August we had bugs attacking it and this was causing glitching.  These issues led Jim & TJ to make a diagnostic to flag BRS system when it has issues (this is notified me there was something wrong with the BRSx).

On the BRSx Overview we have "BAD" State showing along with a red box around BRS ETMX Velocity.  Since we have issues with BRSx, will run H1's SEI_CONF in the WINDY_NOBRSX state until this can be addressed during the day by Jim/Hugh.  Here are some attachments with info on BRSx:

  • 1st attachement is a look at BRS  entering a NOISY state and railing at +/- 30k.
  • 2nd attachment is a look a the BRS Health scope after an hour.
  • 3rd attachement is a screenshot of the BRSx Overview (when it was flashing a "Bad" state notification)

Notes on EX SEI:

After the lockloss and the Watchdog trips, the ISI would trip immediately after untripping and letting Guardian attempt to take the ISI to FULLY ISOLATED.  So I used instructions Jim gave me when I had similar issues recently.  Basically, I worked with individual SEI guardian nodes first:

  • Stage1, State2, & SEI guardian nodes taken to INIT.
  • Untripped Watchdog.
  • Let SEI master node take SEI to FULLY ISOLATED.
Images attached to this comment
jim.warner@LIGO.ORG - 08:58, Monday 30 September 2019 (52202)

Summary pages are down and my remote connection is hopeless, but Corey's image shows BRSX clearly going out of range on his screenshot of the BRS overview (the lower right drift trace shows smooth curves alternated with discontinuous flat sections, doesn't look like real motion). When I checked driftmons before the weekend, BRSX still hand 4-5000 count margin. I suspect that VEA temps have driven it out of range, as a reaction to the cold over the weekend.