At 05:50 PM 1/22/98 -0500, Brian H. Toby <toby@rrdhora.nist.gov> wrote:
>Chuck,
>
>In short, don't use that function it does not work well for even minor levels
>of assymetry. The assymetry function to use is the Finger-Cox-Jephcoat
equation
>[L.W. Finger, D.E. Cox & A.P. Jephcoat (1994) J. Appl. Cryst.,27,892-900.].
>That function has three input values, (sample height, detector height &
>instrument radius) all of which you should know, and thus has no terms to
fit.
>It works well. Don't bother with split-Pearson functions either. They have no
>physical basis.
>
>My guess is that if you send an e-mail message to Larry Finger
>(finger@geolab.ciw.edu) he will send you a subroutine in the computer
language
>of your choice. Well, maybe not COBOL.
>
I agree with everything Brian said. My routines are available at
ftp://cryst.ciw.edu/profile with source code in C, Fortran and Pascal. On
that site are functions that calculates the convolution of the axial
divergence asymmetry with a pseudo-Voigt profile and a small driver program
to show its usage. Also included are output files showing the 'correct'
results, so you can check your compiler, etc.
These routines are further described in a forthcoming 'Computer Program
Abstract' to be published in J. Appl. Cryst.
One caveat - the routines are NOT appropriate for Guinier data in their
present form because the sample to crystal distance is not constant. I have
the necessary corrections and will send them to anyone who requests them.
Larry
--Larry W. Finger finger@gl.ciw.eduGeophysical Laboratory Phone: +1 (202) 686-2410 X 24645251 Broad Branch Road N.W. FAX: +1 (202) 686-2419Washington, DC 20015-1305, USA http://granite.ciw.edu/~finger/ http://btgix8.bgi.uni-bayreuth.de/~lafi