Properties of collimators

[Martin Meven]

Dear Kristian Nielsen,

thank you four your quick answer.

I would like to be more specific about my question. For our instrument
within the FRM-II project I want to use collimators on the primary beam
to reduce background and radiation at the monochromator hole in the
biological shielding. Furthermore I would 
like to estimate the amount of neutron radiation at the position of the
crystal sample in dependence of collimation.

The geometry of our beam line is quite simple:


             REACTOR           #      BIOLOGICAL SHIELDING
            "guide nose"       #                             #
            |            collimator            monochromator #
source -------------------=====---------------------\        #
       < d1>|< d2              ><   d3             > \       #
       A                       B                    C \ d4  # 
                               #                       \   #
                              
#########################\##                                                         
#                         \
                               #                          \
                                                     D   sample

Distances         | Beam width (w x h)
-------------------------------------
d0=0 mm           | flat source about 10^9 n/s/cm^2, infinite plane
d1=1780 mm        | 70 x 100
d2=4300-1780 mm   | 50 x 90
d3=6690-4300 mm   | 40 x 100 Cu-Monochromator, reflection angle Theta
20,30,40
                  | degree
d4=1300+500 mm    | 50 x 50, sample size max. about 10 x 10 mm

I think that an approximation using a soller component should be
sufficient.

The beam tube is slightly konical and has an inherent collimation due to
its
length of about 1 degree. 

To calculate the distribution and intensity at the monochromator and the
sample position I used a monochromator divergence of 0.2 degree.
I got from this as a rough estimation a reduction of the overall
intensity at the
monochromator area of about 43% using a 30´ collimation and allmost no
effect at 
the sample position. 
This calculation was done by simply calculating the divergences in the
diffraction
plane without taking into account different wavelengths or the vertical
divergence.

We can use three different primary collimations due to the shutter
system in the
reactor wall. Therefore we have to decide which collimations are useful.
Usually
the collimations 60´, 30´, 15´ are used. In this case I think the 60´
collimation 
can be dropped. But as you can see, my calculations are quite
"handwaving". 

Therefore to check whether my thoughts are reasonable I would like to
discuss the
matter with you. If there are further questions or suggestions please do
not 
hesitate to call or write me.

With kind regards

Martin Meven