Data Set Overview
  =================
 
  This data set includes all data from the High Rate Detector (HRD)
  through the end of 2006.  All data products in this data set are in
  tabular format. Since HRD is mounted on the same platform as the
  Dust Analyzer, the pointing is the same for both instruments.
 
  Data file types and contents
  ============================
 
  Data files in the 'raw' subdirectory contain all recorded events,
  including impacts, noise events, and calibration events.  The first
  column of each file is a unique event number of the format
  Xnnnnnnnnn where X is a code for the year (A = 2000, B = 2001, etc,)
  and nnnnnnnnn is a sequential number within the year.  The raw
  filenames are of the form hrd_yyyy_doy_doy.tab and include data
  within the date range specified.
 
  Data files in the 'calibrate' subdirectory contain calibration
  events extracted from the raw files.  Note that the sample rate is
  changed periodically, resulting to changes in the time increments in
  the calibration files.  Calibration files are grouped by year, and
  since in-flight calibrations were not done in all years,
  subdirectories do not exist for all years of the mission.  The event
  number corresponds to that in the raw file.  The filenames are of
  the form hrd_yyyy_doy_doy_cal.tab and include data within the date
  range specified.
 
  Data files in the 'processed' subdirectory have had calibration
  events and noisy events removed.  On 2005-248 there occurred an M4
  event in the large detector which resulted in a noisy M1 threshold.
  Since then, all M1 events are considered to be noise and have been
  removed from the processed data unless the M2 threshold is triggered
  or the small detector m1 threshold is triggered or the large
  detector High Mass is set.  The processed files have filenames of
  the form hrd_yyyy_doy_doy_prc.tab and include data within the date
  range specified.
 
  The calibration and processed files include a quality code flag.
  The QC flag indicates a discrepancy in the data line and is set if
  any one of the following obtain:
      1.  Missing latch data.
      2.  Counters change with no latch data.
      3.  Higher threshold triggered and lower threshold did not.
 
  The data directory 'onoff' contains files with the on-off times of
  the HRD, divided by year.  These have filenames of the form
  hrd_yyyy_on_off.tab. Note that if the instrument was not powered on
  or off during the covered time period, there will not be an on-off
  file for that time period.
 
  The data directory 'pointing' contains hourly instrument pointing
  and spacecraft positions generated from the SPICE kernels.  The
  pointing data pertains to both the CDA and HRD instruments, which
  are mounted on the same platform.  Gaps in the pointing and position
  data are due to the fact kernels are not available for those time
  periods.  The metakernel listing the kernels used to generate the
  pointing and position files is included in the document directory.
 
  Calibration of the HRD Detectors
  ================================
 
  The HRD calibrations are similar to those for the Dust Flux Monitor
  Instrument(DFMI) on the STARDUST misssion to comet Wild 2.  The
  calibrations of HRD detectors for the Cassini mission as well as for
  those for the DFMI have done with dust particle accellerators in
  Heidelgerg and Munich.  During the calibration at Heidelberg iron
  particles in the velocity range of 1-12 km/sec were used while at
  Munich accelerator glass particles at similar range  were used.
  Please refer to Simpson and Tuzzolino (1985), Simpson and Tuzzolino
  (1989), and Tuzzolino et al. (2000) for more details and description
  of how these calibrations were performed.
 
  Besides laboratory calibration, in-flight pulser calibrations were
  performed periodically to ascertain the performance of the
  electronic system of the HRD instrument.
 
  References
  ==========
 
  Srama, R., T.J. Ahrens, N. Altobelli, S. Auer, J.G. Bradley, and 35
  others, The Cassini Dust Analyzer, Space Science Reviews 114,
  465-518, 2004.
 
  Simpson, J.A. and Tuzzolino, A.J., Polarized Polymer Films as
  Electronic Pulse Detectors of Cosmic Dust Particles, Nucl. Instrum.
  Methods A236, 187-202, 1985.
 
  Simpson, J.A., D. Rabinowitz, and Tuzzolino, A.J., Cosmic Dust
  Investigations, I:  PVDF Detector Signal Dependence on Mass and
  Velocity for Penetrating Particles, Nucl. Instrum. Methods A279,
  611, 1989.
 
  Simpson, J.A. and Tuzzolino, A.J., Cosmic Dust Investigations, II:
  Instruments for measurement of particle trajectory, velocity, and
  mass, Nucl. Instrum. Methods A279, 625, 1989.
 
  Tuzzolino, A.J., R.B. McKibben, J.A. Simpson, J.A.M. McDonnell, M.J.
  Burchell, and 8 others, Calibration of the Dust Flux Monitor
  Instrument (DFMI) for the Stardust Mission to Comet Wild 2, in the
  Stardust Docushare, Jet Propulsion Laboratory, 2000.

Confidence Level Overview
  =========================
    Large impact during Ring Plane Crossing
    =======================================
 
    The HRD performed well throughout the entire mission. However,
    after crossing the Saturn G ring on DOY 2005_248, the #1 large HRD
    sensor was hit by a large size dust particle that exceeded all
    mass thresholds of the HRD instrument (the size was greater than
    several hundred microns). This event probably changed the
    capacitance of the sensor and the HRD instrument started to show
    some noise in the lowest threshold M1 of the #1 sensor in the
    range of several counts per mimute. It was decided not to change
    the threshold level by a factor of 10 (the only possible response)
    in order not to lose the data from the other three thresholds (M2,
    M3, M4). So, the M1 data are not trustworthy for low count rates,
    but they could be useful when the counting rate is high during the
    ring crossing, or satellite flybys.
 
    Both detectors triggered
    ========================
 
    In some events in the 'raw' data, both detectors are triggered.
    This implies simultaneous impacts in the two detectors.  In the
    'processed' files, such events have been split into two lines, one
    for each impact. Both lines have the same event number since they
    come from the same entry in the corresponding events file.
 
    Note that within a single detector, the triggering of a higher
    threshold will ordinarily be accompanied by the lower thresholds
    being set also. If not, the impact is suspect and the quality code
    is set.
 
    Cumulative count discrepancy
    ============================
 
    In some instances the cumulative counts don't agree with the
    individual counts numerically.  This is due to the processing time
    the CPU takes to reset the latch for the next discrete event.  If
    an event occurs during the latch reset cycle the event will not be
    saved by the latch and the CPU will read a zero value.  However,
    the missing latched event will still be recorded by the eight
    16-bit counters.
 
    HRD instrument reset
    ====================
 
    On instrument power-on, the HRD performs a power-on reset.  The
    reset sets the HRD clock and all the counters to zero.  Since the
    counters are set to zero on reset, the counts appear to go down
    with time on reset. Missing data can be determined by looking at
    the on-off files (described above) which contain a list of the
    power on and off times.