Dear Josh, Just in case some more clarification is needed:
the problem arises because of parameter correlations: the (2-theta independent) zero shift and parameters describing angle dependent shifts (displacement and transparency) are very strongly correlated. So, simple advice: never refine them together. Choose either of two possibilities and in 99.9% cases it would be sufficient. In GSAS/EXPGUI you will not be able to refine those parameters simultaneously: it is explicitly forbidden by the EXPGUI author exactly to avoid this kind of problems. If you believe that you have good data quality and simultaneously can clearly recognize two types of shift, refine sequentially and (you can do this in Fullprof) set as low shift (via parameter code) as possible - this would make the refinement a bit more stable. Sincerely, Maxim. _________________ Maxim V. Lobanov Department of Chemistry Moscow State University loba...@icr.chem.msu.ru *************************************** This is a forwarded message From: Yokochi, Alexandre <alexandre.yoko...@oregonstate.edu> To: Lubomir Smrcok <uachs...@savba.sk>, Josh Kim <joshkim....@gmail.com> Date: Thursday, January 15, 2015, 2:16:03 AM Subject: FullProf ===8<==============Original message text=============== If I recall it correctly (haven't had the opportunity to play with Fullprof for a few years) Sycos is an instrument misalignment/sample transparency correction parameter. Perhaps you are using an impossible initial value for it - should be very close to zero. As I share with my students on occasion, RTFM (at least for instances of F = fantastic). Dr. Y -----Original Message----- From: rietveld_l-requ...@ill.fr [mailto:rietveld_l-requ...@ill.fr] On Behalf Of Lubomir Smrcok Sent: Wednesday, January 14, 2015 2:05 PM To: Josh Kim Cc: rietveld_l@ill.fr Subject: Re: FullProf well, do you know why are you refining just that parameter ? lubo On Wed, 14 Jan 2015, Josh Kim wrote: > Dear Rietvelds, > > I am new to FullProf, and am familiar with the difficulty in preparing a > decent .pcr file, but have run into a problem to which I cannot pinpoint the > error: During the first iteration of refinement I get the following message: > > "Singular Matrix!! Problem with SyCos_pat1 no.: 1? > > Any ideas? I can provide the .dat and .pcr file if anyone is interested > > Thanks > Josh ===8<===========End of original message text===========
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Please do NOT attach files to the whole list <alan.he...@neutronoptics.com> Send commands to <lists...@ill.fr> eg: HELP as the subject with no body text The Rietveld_L list archive is on http://www.mail-archive.com/rietveld_l@ill.fr/ ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
