Readme for Indra2 macros set. 8/20/2007 release 1.61 Jan Ilavsky, ilavsky@aps.anl.gov ************************************ This set of macros is developed for evaluation of USAXS data obtained on UNICAT 33ID beamline instrument. It runs on Wavemetrics Igor Pro version 6 and higher. Revision history: Indra 2 set. __________________________________________________________________________________________________ August 20, version 1.61 Removed bug when MSAXS correction failed after user used different tool and changed folder. Modified Desktop loader for David Londono - added option to use file name as sample name and changed the selection method to allow easy multiple file selection with shift-click. Made compatible with Igor 5.05 and higher. (MSAXS graph fix) Removed bug when MSAXS graph was showing at wrong time. Fixed position of the main graph and main panel. July 15, 2007, version 1.60 Main release of the new USAXS reduction procedure. Still Only Igor 6.0 compatible. May 17, 2007, version 1.51 Fixed bug in Rigaku/Osmic data support created by new reduction panel. April 17, 2007, version 1.50 Added whole new main panel in which reduction is doen in one step (more or less). Same panel used for Blank and sample. Resulted in some changes to wave names and at this time may be breaking some older code functionality. Also, only Igor 6 compatible. October 18, 2006, version 1.42 Changed dialog which appears when user does not save R data and wants to continue to more easy to understand and easie to use... Changes place where Indra stores temporary files to allow users to run as no-privileges user who has no rights to write into Program files folder. Add button to reload same spec file to Spec import panel. August 2006, version 1.41 Patch for non linearity of the USAXS diode system observed in AUgust 2006. Will check if needed and if so, it will put out dialog with the correction factor. Default should be OK, but for details talk to me... April 7, 2006 version 1.40 Desmearing - added some more controls to 1. load Irena desmearing (if present), 2. load Irena package (if present in Igor but not loaded) and start desmeraing, 3. Tell user where to go for the Irena package. Included beta version of loader data for Osmic/Rigaku Desktop USAXS instrument. For proer function need new data format, but should work with limitations with the old one too... Few minor changes to improve stability of displaying peak on lin-lin plot in Subtract Blank from Sample Removed Autofit Gauss - did not work... 1.30a and 1.30b Needed to make minor changes due to changes in the IN2_generalProcedures package. Attempting to do better in peak height/width calculations. Not much improvement... Modified subtract sample and blank routine to avoid any chance of user accidentally leaving the sample thickness 0 mm, which was causing lately problems. Added I0 dark current in the calculations. It is now available, let's use it. December 1, 2003, release 1.30 Redesigned Spec import data tool to use panel, not old style dialogs. Should avoid some user confusions. See the manual for use. Fixed minor bug in Plot with Math, which caused problems when graph did not exist and user tried to change their setting. Fixed minor bug in convert Raw to non-USAXS, which caused problem with generation of folder names under some conditions... October 23, 2003, release 1.27a Update of spec.ipf. Add requested feature. No other changes. October 23, 2003, release 1.27 Minor update: Moved all menu items to one, use submenus. Should save some space for users on the screen. Fixed desmearing bug, which caused crash when user dropped cursor on wrong wave. Minor patches. May 12, 2003, release 1.25 Minor update: Minor fix for AnisoScans used in SBUSAXS. Now scales the data to range 1 (* 10^8). This is so the data from different aniso scans can be compared together. Note, no attempt to scale for other calibration parameters is attempted... January 1, 2003, release 1.24 Minor updates: Desmearing update for export of waves from within the routine. Previous code failed with data different than DSM, new fix should export also M_DSM data and any name data. Minor modification of panel where user selects the spec wave names. Modification of Standard Plots, which should now allow displaying data types of differnet types... Sorrely needed function. October 18, 2002, release 1.23 Few minor updates and bug fixes. Nothing major. Removed Sizes part and moved to Irena 1 where it is with new GUI. March 1, 2002, release 1.20 Larger change so skipped few numbers here... This is update, in which I had to make General Procedures incompatible with both Indra2 and Irena 1 previous versions. Combination with old Irena 1 and Indra 2 will not compile... Therefore users of both packages MUST update both at the same time - now... Some usual bug fixes: 1. Major change in desmearing routine, removed "fast" method and modified parts of the extrapolation, made input using cursors... 2. Added some features in R wave creation. Fixed few bugs, which were not too bad, but annoying. January 30, 2002 release 1.17 Maintanance release, minor change. Fix for yet unused piece of code - merging different data sets. Minor change with look of one graph (R_wave creation). December 8, 2001 release 1.16 Maintanance release, minor change - Spec file now may include sample thickness. Igor code now check, if this value is included and if so, it will offer it as sample thickness. No problem, if the value does not exist - 1 is still default value. October 25, 2001 release 1.15 Major redesign of the errors for the R_Int... This seems to be very close to the useful method (propagation of poisson errors) + 1% of UPD intensity as constant error... The final errors for SMR data are up to 5-6% at the high Q region, which looks OK. Now the fit for beam center in R_wave procedure is weighted by errors as well as the errors from the fit are propagated forward through wavenote and eventually are used to calculate the Kfactor error (from the Blank FWHM error and transmission error). Minor change in behavior of the importing data from Spec to Igor. More automated for USAXS users... Also added verification, that the wave names in spec exist for each scan which is being converted... I wonder how much checking needs and can be done... Added new MSAXS evaluation procedure with some more needed functionality. Allows not to weigh the results for caculation of average transmission of the sample and allows to have sectors selected, which are recalculated. Little housekeeping willbe necessary at some point, but should be sufficient at this time. Added (by request) also manual overwrite of the average transmission. September 23, 2001 Intermediate change in way we calculate errors for R, SMR and for SB-USAXS also DSM waves - now we use 1% of the R Intensity is the major component of the error - we have found out from experiments, that this is apparently best error estimate. We add first principal error calculations, but these are negligible. For now we neglect increase of error in the tails at high Q region... Added feature requested by Pete - Size distribution now offers user to recover old fitting parameters, if it finds out, that the size distribution was already run on the data... The logic now is about this: User has chance to recover already used parameters, if decides against it, the parameters are set to the last used in this Igor experiment, and they do not exist, they are set to default values... Few minor cosmetic changes. September 1, 2001 Added calculation panel for size distributions. Will work only when displayed is Size Distribution (volume f(D)). Note, that the math was checked but need to be treated with caution - for now, before I have chance to verify formulas. Function with formulas added into the General Procedures. Subtract sample from blank now asks for restarting with new sample if no SMR (DSM for SBUSAXS) data were created. This prevents users from clicking buttons too fast mistake... Sample thickness is now written in the wavenote for each wave. Further, the sample thickness is carried forward in this order: first is set to 1([mm]), then set to previous sample thickness(if exists), then set to previously used thickness for this sample from wavenote (if exists), and then presented in dialog to user. Fix problem when subsequent running of Raw-to-USAXS second time on the same data caused loss of SDD from the list of parameters, which caused SMR wave to be full of NaNs, when run without R_wave prior. This was obscure bug, but appeared. ANyway, now the SMR wave procedure, if the SDD is missing reruns function which fixes this. It is still unadvisable to rerun Raw-to-USAXS as I am not sure about the consequences. ___________________________________________ August 1, 2001 Initial release was actually 1.1 version of most macros. Release 1.0 was distributed on limited scale. Release 1.1 is first which seems to contain all pieces of the code in working order and was released in early August 2001. This is the first release, which uses version numbers to guarrantee the compatibility. ___________________________________________ Indra 1 set. this set is currently retired and was used from year 2000 until January 2001. I strongly suggest that, if you have data files using these macros you keep the macros safely, so you can access the data. version 1 and 2 of macros are incompatible. I have developed function which converst the old (Indra 1) data into new (Indra 2) schematics, but it is not perfect and is so difficult to use, that I suggest to everyone reevaluation of data with the new macros. Old data sets should be readable without any problems.