Reports until 14:52, Tuesday 08 September 2015
H1 AOS
betsy.weaver@LIGO.ORG - posted 14:52, Tuesday 08 September 2015 - last comment - 18:19, Tuesday 08 September 2015(21304)
HARDWARE INJ OFF

Note, when the EXC bit in the CALCS CDS overview is in alarm, we tend to open the screen CAL_INJ_CONTROL to attempt to diagnose - This shows a big red light for some ODC Channel OK Latch, leading us to misdiagnose what is actually in alarm.  We have 2 operational problems:

 

1) If generically, there is a red light on the CDS screen - where do you go?  Normally, we follow the logical medm and are able to get to the bottom of the red status via logical nested reds.  This is not the case for the CALCS screen - the CDS H1:FEC-117_STATE_WORD bit is RED on the H1CALCS line of the overview screen, yet this bit is nowhere on the CALCS screen.

So, where does the info come from for specifically the EXC bit of the H1CALCS state word, such that we can do something about it?

 

2) Someone should rework the CAL_INJ_CONTROL.adl so that it doesn't cause us to misdiagnose actual reds.  Currently, the HARDWARE INJECTIONS are out of configuration (outstanding issue to still be sorted) and yet, there is NO INDICATION of that on the CAL_INJ_CONTROL screen...  Also, the CW injection appears to be off, but there is no "red alarm" on the screen.

 

BTW, the HARDWARE INJ appear to be off.  They dropped around 7pm local time last night (20 hours ago).

Images attached to this report
Comments related to this report
betsy.weaver@LIGO.ORG - 14:52, Tuesday 08 September 2015 (21305)

Images attached to this comment
jameson.rollins@LIGO.ORG - 15:21, Tuesday 08 September 2015 (21307)

The hardware injection folks should comment, but I spoke to Duncan Macleod at LLO last week about redesigning that screen.  I told him not to use red for things that don't actually indicate errors, or conditions that need to be acted upon.  I think he was working on the screen, so maybe it can be updated at LHO soon.

The EXC bit in the CALCS is expressly excluded from the DIAG_EXC checks, because excitations in the CALCS model are used for hardware injections.

jenne.driggers@LIGO.ORG - 18:19, Tuesday 08 September 2015 (21313)

Our confusion with the Cal EXC was that we were expecting an excitation to exist, but it didn't.  The CDS overview screen that Betsy posted is modified to show red for the CALCS EXC when there *is not* an excitation, whereas it shows red for every other model if there *is* an excitation.  But, we were having trouble on the CAL CS overview screen determining if there was actually a problem.