Date: Wed, 27 Jun 2001 13:13:28 +0200 (CEST)
From: Stuart Rycroft <rycroft@iri.tudelft.nl>
To: Ulrich C. Wildgruber MPI fuer Metallforschung Stuttgart
<wildgrub@dxray.mpi-stuttgart.mpg.de>
Subject: Re: SiC.comp
Hello
I have just run the sample using a simple instrument template. The
attached "reflectometer.instr" file is a reactor source with a
monochromator, two slits, sample and detector.
The "control.bat" file shows the variable settings to produce the
reflectivity profile:
samp_rot - Sample rotation (90 degrees is parallel to the beam)
det_lim - Since the reflected beam position changes with the sample
rotation, you can set the minimum spacial position of the
detector. This ensures that the reflected beam does not miss
the detector.\
x_slit - The width of the slits. This is determined by the size of
footprint on the sample. I usually keep to less than
1cm. This can be increased for the larger angles.
x_pos_fact - Since the sample is positioned from its edge not its
center it is necessary to move it otherwise it will only
cover half the incident beam. This variable should be
altered if the slit width is increased since you then have
a bigger beam.
I also ran the simulation without the sample - "no_samp.instr" to get a
direct beam run so that I could work out the reflectivites.
Also attached is "SiC.comp"
I attach the results in the next mail.
Stuart
P.S the run times are rather long (ncount=5e8) You can reduce this
if you open the slits more for the higher angles. I kept the slits the
same since I only did one normalisation run (i.e. without sample)
Attached are the detector files plus an excel file showing the plot. I
have simply summed the counts on the detector for each angle and then
compared this to the normalisation run to get the reflectivity.
I have currectly left the simulations of the ROG reflectometer. It was
taking too much time and we needed to move on with the project for which
we were funded namely the polarised neutron part of SCANS.
However I have a complete setup for ROG which required a chopper
(already in mcstas) but also a supermirror guide and frame overlap mirror
inside a guide. These had to be developed. So the simulation is
complete. The time problem came when we had to compare the simulations to
real results and so we have put this on hold.
Anyway, I guess you will see that the SiC samp is quite straight
froward...just uses Rayleigh to change the probability.
If it is useful to you then let me know if you need/want further
explanatins of my method.
Bye for now
Stuart Rycroft
Interfaculty Reactor Institute (IRI)
Delft University of Technology
Mekelweg 15
2629 JB
Delft
Netherlands
Tel: (+31) 015 278 3900
Legal notice