Ignore:
Timestamp:
Jul 17, 2018 1:58:12 PM (4 years ago)
Author:
srkline
Message:

two significant changes:

1- in the averaging, before the data is finally written out, duplicate q-values (within 0.1%) are averaged so that duplicate q-values do not appear in the data file. This has a very bad effect on the calculation of the smearing matrix (dq=0).

2-for the data from the back (high res) detector, two steps have been added to the processing at the point where the data is converted to WORK. First, a constant read noise value of 200 cts/pixel is subtracted (found from average of multiple runs with beam off) Second, a 3x3 median filter is applied to the whole image to eliminate the stray bright pixels. Some are saturatd, some are simply outliers. Very effective.

File:
1 edited

Legend:

Unmodified
Added
Removed
  • sans/Dev/trunk/NCNR_User_Procedures/Reduction/VSANS/V_WorkFolderUtils.ipf

    r1109 r1111  
    464464//(the function Add_Raw_to_work(type) adds multiple runs together - and is LOW priority) 
    465465// 
     466// JUL 2018 
     467// now removes a constant kReadNoiseLevel from the back detector (this is a constant value, not time 
     468//  or count dependent). It does not appear to be dependent on gain, and is hoepfully stable over time. 
     469// 
    466470//the current display type is updated to newType (global) 
    467471// 
     
    502506// each correction must loop over each detector. tedious. 
    503507 
     508        //except for removing the read noise of the back detector 
     509        NVAR gIgnoreDetB = root:Packages:NIST:VSANS:Globals:gIgnoreDetB 
     510 
     511        if(gIgnoreDetB == 0) 
     512                Wave w = V_getDetectorDataW(fname,"B") 
     513                w -= kReadNoiseLevel            // a constant value 
     514                 
     515                MatrixFilter /N=3 median w 
     516                Print "*** median noise filter applied to the back detector***" 
     517        endif 
     518         
     519         
    504520        // (0) Redimension the data waves in the destination folder 
    505521        //     so that they are DP, not integer 
     
    819835        // put this in some separate work folder that can be cleaned out at the end (ADJ) 
    820836        String tmpType="ADJ" 
    821         V_Raw_to_Work(tmpType) 
     837         
     838        //this step removes the read noise from the back so that neither added file will have this constant 
     839        V_Raw_to_Work(tmpType)           
    822840                         
    823841        //now make references to data in newType folder 
Note: See TracChangeset for help on using the changeset viewer.