Last week during a low-noise lock the COMM tidal signals were hitting their software limits.  This is a somewhat insidious problem: right now we don't have a good monitor for these limits, and there are some channels (like the tidal drives) that can hit these limits without causing the IFO to lose lock.

I've written a python script that will scan the frame files for any channels that have their LIMIT enabled, and check whether that limit was reached over a span of GPS times.  The script is not elegant, and I'm hoping someone from DetChar can modify it and make it faster.  For now it provides some important information: which channels to look at, which channels to ignore, and how to figure out if the LIMIT is enabled from the SWSTAT setting.  (Turns out it's the 13th bit.)  Note that the integrator filter module that is used in the tidal drives and the ALS WFS has a different channel to indicate whether the limit is enabled.  Gotcha!

The script is on the LHO cluster here, the current version only runs on the LDAS clusters on archived data (~10min old or more), I will write a version that can run in the control room using cdsutils.  It's not fast, mostly because of the frame-reading.

For H1, there are currently 107 channels with their LIMIT enabled during low-noise, a list of these channels with their LIMIT value is attached.  (Some of these aren't important, like the ITM oplev outputs.)  For the lock on Apr 2 there were three channels that hit their software limit: the two ETM common tidal drives (really, the same signal), and the ETMX BRS output, which wasn't being used.