[Date Prev][Date Next][Thread Prev][Thread Next][Search] [Main Index] [Thread Index] [HEASARC Mailing List Archives]

ROSAT Status Report #149: SASS/HRI Changes



  Date: Mon, 24 Feb 1997 10:25:17 -0500
  Subject: ROSAT STATUS REPORT #149: SASS/HRI Changes
  
  .....................................................................................
  
  
  	 REPORT ON SASS/HRI CHANGES RELATED TO THE HIGH BACKGROUND THRESHOLD
  
  
  1.0 CHANGES TO THE CUTOFF THRESHOLD
  2.0 BUG IN THE AVERAGE BACKGROUND CALCULATION
  3.0 CHANGES TO FIELD SUMMARY PAGE
  
  
  1.0 CHANGES TO THE CUTOFF THRESHOLD
  
  	G. Hasinger notified MPE that significant amounts of his data
  were rejected because of 'high background', even though they were
  acceptable.  It was found that the SASS/HRI algorithm which calculates
  the high background threshold (above which data are 'rejected')
  occasionally produced quite small values for long observations with
  mostly very low background levels.  Because significant amounts of
  acceptable data were rejected, the target often had to be rescheduled
  for further observations when if fact, the proposal's objectives had
  already been met.  Therefore, MPE requested a change in the algorithm
  to remedy this situation.
  
  	The change in the method of determining the cutoff level for
  the 'HB' parameter (the background rate which appears in the 'Time
  History' section of the SASS output) is to force the cutoff so as to
  accept at least all time segments with HB <= 8 (E-7 c/s/pix).  The
  previous algorithm (which determines the cutoff by finding the HB
  level for which the s/n for point source detection begins to get worse
  with the addition of higher background data) remains unchanged for
  higher levels.
  
  	This change has been implemented with SASS7_9, HRI6_3_3 and
  the output also can be identified by the presence of additional
  background parameters (see below). We have examined the output and
  source lists for various test sequences and all parameters appear to
  behave as expected.  Our evaluation of the source lists for the test
  runs indicates that alterations are minimal and will not impact users
  adversely.
  
         [NB:     HB is the background level in units of E-7 c/s/pix.
                  HB is always an integer, with typical values of 3 to 5, 
                  occasionally  > 10, and on rare occasions > 50.]
  
  2.0 BUG IN THE AVERAGE BACKGROUND CALCULATION
  
  During testing for the change in background rejection threshold, we
  discovered a bug in the calculation of the 'average background rate in
  field' (which is given in the Observation Summary).  The bug (which
  affects only this parameter and has no further consequence) is that
  the counts are accumulated FOR ALL TIME INTERVALS, including those
  which are rejected because HB exceeds the cutoff; whereas the time
  used to compute the average is the ON time of only the accepted data.
  
  3.0 CHANGES TO FIELD SUMMARY PAGE
  
  In addition to fixing this bug, we now calculate and print the values
  of other parameters which can assist the user in evaluating the
  usefulness of data rejected because of high background.  Here are
  examples of the previous and new versions.
  
  
  
  PREVIOUS
  
  Observation Summary               :
   
   The MicroChannel Plate gain between 3' and 6' off-axis for this observation is  5.79
   which should be between 5.  +/- 2.0
   Average background rate in field:    0.0038786 ( counts/sec/arcmin**2 )
   Total accepted counts in field:        10189
   
  
  NEW 
  
  [NB: This is a different observation than the preceding one]
  
  Observation Summary               :
   
   The MicroChannel Plate gain between 3' and 6' off-axis for this observation is  4.87
           NB: This value should lie within the range 5 +/- 2.
   Total accepted counts in field:                                                       188628
   Highest accepted background level, forced to be >= 8. (E-07 counts/sec/pixel):            8.
           NB: 1E-7 counts/sec/pix corresponds to 0.00144 counts/sec/sq_arcmin
   The total time rejected because of a high background rate (secs):                    1850.00
   Average background rate in field (counts/sec/arcmin**2) for accepted data:         0.0044007
   Average background rate for times rejected because of high background              0.0582850
           NB: '-1' means no data have been rejected because of high background
   Average background rate in field if no high-background data were rejected:         0.0073638
   
  
  24 February 1997				D. E. Harris, for RSDC/SAO
  **************************************************************************
  Internet - rsdc@cfa.harvard.edu		RSDC    MS-3
  					Center for Astrophysics
  UUCP	 - ...!harvard!cfa!rsdc		60 Garden St. Cambridge MA 02138 USA
  					tel: (617) 495-7134  FAX: 495-7356
  **************************************************************************