magGlitch

Magnetic Field Glitch Monitor

magGlitch internal constants and functions

magGlitch

magGlitch checks magnetometer channels for glitches within relatively broadband noise. Due to the many peaks in the magnetometer spectra which tend to dominate the band-limited power, bands are selected which include as few peaks as possible. Glitches which occur less than .1 seconds apart are considered the same glitch.
Usage: magGlitch [<conf_file>]

Output
Generally, no output is rewritten unless the monitor is started more than once in the same day. Instead, old log filenames and data filenames get the current date appended to them.

Log File-- The log file is usually located in the DMTHTMLOUT directory for magGlitch. On the web this can usually be viewed by clicking on the magGlitch link from the DMT Monitor Observer web page. At LHO, this is http://blue.ligo-wa.caltech.edu/gds/monitor_reports/magGlitch/ and at LLO, this is http://london.ligo-la.caltech.edu/gds/monitor_reports/magGlitch/. Log files report the specifications made in the configuration file, the number of glitches found each Log Interval (both described below), and--- at the end of the run--- a histogram for each band indicating the total number of glitches found, binned by peak magnitude in units of .

Data File-- The data file is usually located in the DMTOUTPUT directory for magGlitch, and contains the records of all glitches generated. Data records are specified in the configuration file, described below. There are two types of records:

• glitch record- A single-line record, separated by spaces, with the following information:
  chan_name GPS_start GPS_end GPS_peak peak_mag ts_avg ts_sdev npts com
  where chan_name is the name of the channel, GPS_start/end/peak are the start, end, and peak times of the glitch, peak_mag is the peak magnitude of the glitch (in ), ts_avg/sdev are the current values for the running mean and standard deviation of the full time series, npts is the number of points exceeding the threshold during the glitch, and com is an optional comment field to describe the type of glitch which occurred.
• histogram record- A multi-line, formatted dump of the histograms generated during the previous Log Interval (described below). A histogram is generated for each band observed, binning the glitch peak magnitudes.

Configuration File-- The configuration file sets values for run-time parameters and also provides the list of channels to run on. A sample configuration file follows:

 DataFilePath magE5H_200-300.dat
 LogFilePath magE5H_200-300.log
 LogInterval 3600
 DAQInterval 84600
 DecayTime 10.0
 TriggerON 0
 
 Channel H0:PEM-BSC1_MAG1X 2048 2.0 2.5 BandPass 5 185.0 235.0 185-235 End
 Channel H0:PEM-BSC1_MAG1Y 2048 2.0 2.5 BandPass 5 245.0 265.0 245-265 End
 Channel H0:PEM-BSC1_MAG1Z 2048 2.0 2.5 BandPass 5 280.0 293.0 280-293 End
 Channel H0:PEM-BSC1_MAGX 2048 2.0 2.5 BandPass 5 185.0 235.0 185-235 End
 Channel H0:PEM-BSC1_MAGY 2048 2.0 2.5 BandPass 5 245.0 265.0 245-265 End
 Channel H0:PEM-BSC1_MAGZ 2048 2.0 2.5 BandPass 5 280.0 293.0 280-293 End
 Channel H0:PEM-BSC10_MAGX 2048 2.0 2.5 BandPass 5 222.0 295.0 222-295 End
 Channel H0:PEM-BSC10_MAGY 2048 2.0 2.5 BandPass 5 222.0 295.0 222-295 End
 Channel H0:PEM-BSC10_MAGZ 2048 2.0 2.5 BandPass 5 222.0 295.0 222-295 End
 Channel H0:PEM-COIL_MAGZ  2048 2.0 2.5 BandPass 5 207.0 232.0 207-232 End
  

An explanation of the above parameters:

• DataFilePath - Data output filename
• LogFilePath - Log output filename
• LogInterval - Number of seconds between log output
• DAQInterval - Number of seconds to run the monitor. If not set, or equal to 0, magGlitch will not stop unless killed.
• DecayTime - Coefficient of exponential decay for the running averages and standard deviations.
• TriggerON - Flag whether or not to attempt to write triggers for each glitch.
• Channel - Channel entry line. All specs on the entry line must be separated by a single space. The order of specs are:
  <chan_name> <sample> <lo_thresh> <hi_thresh> <filter> <order> <lo_band> <hi_band> [<com>] End
  where chan_name is the name of channel to be monitored, sample is the sampling rate of the channel, lo/hi_thresh are the two sets of thresholds to be used, filter can be one of LowPass, BandPass, or HighPass, order is the order of Butterworth filter to be used, lo/hi_band are the frequency limits of the filter (only one should be specified when using LowPass or HighPass filters, and com is an optional comment to be placed at the end of each data record for easy identification of glitches.

Suggested Uses
As one can see in the above configuration file, not all channels are set to monitor the same bands. The premise behind this was that to first order, all axes have the same response to a generalized magnetic field transient. Obviously this is not true, but in order to avoid the many dominating peaks in the magnetometer power spectrum and still monitor a relatively wide bandwidth, the magnetometer axes were interspersed between major spectral peaks. In the future, the author plans to allow multiple bands to be monitored for each channel, which would allow axes' configurations to look more symmetrical.

alphabetic index hierarchy of classes

generated by doc++