40m QIL Cryo_Lab CTN SUS_Lab TCS_Lab OMC_Lab CRIME_Lab FEA ENG_Labs OptContFac Mariner WBEEShop
  40m Log  Not logged in ELOG logo
Entry  Mon Aug 24 16:10:07 2015, ericq, Update, SUS, ETMY Oplev laser power is falling ETMYsum_trend.pdf
    Reply  Mon Aug 24 16:46:49 2015, ericq, Update, SUS, ETMY Oplev laser power is falling OLsum_trend.pdf
       Reply  Thu Sep 3 08:53:55 2015, Steve, Update, SUS, ETMY Oplev laser power is falling ETMYoplev2y.png
Message ID: 11526     Entry time: Mon Aug 24 16:10:07 2015     Reply to this: 11527
Author: ericq 
Type: Update 
Category: SUS 
Subject: ETMY Oplev laser power is falling 

Today I noticed the box around the ETMY oplev sum flashing red, as it dipped below 1k. I don't recall seeing this recently, so I wanted to look up the history.

However, we've been having trouble with our minute (and longer) trend data, so I had to hack it out a bit... Here is the unfortunate result:

I think we can be fairly confident that this is not due to alignment drifts, we generally keep the QPD reasonably well centered. I also recentered it today, and the counts remained at ~1k. 

Details of the hack that got me this data:

I ended up looking at the BURT snapshots from every night at midnight, which report a number for ETMY_OL_SUM_OUT16, and making a text file with dates and values with the following BASH spaghetti:

find /opt/rtcds/caltech/c1/burt/autoburt/snapshots/2015 -wholename "*00*/*scy*" |xargs ack --nogroup "ETMY_OL_SUM_OUT16 1" |  sed -e 's/.*2015/2015/g' -e 's/\/c1.*\([0-9]\..*$\)/, \1/g' -e 's/\//-/g'  > ETMYsum.txt

This produces a file full of unsorted lines like: 2015-Aug-23-00:07, 1.106459228515625e+03

The python package pandas is good at parsing dates and automatically plotting time series: 

olsum = pandas.read_csv('ETMYsum.txt', index_col=0, parse_dates=True)
Attachment 1: ETMYsum_trend.pdf  16 kB  | Hide | Hide all
ELOG V3.1.3-