Running INTERFERE for GEM Data  

INTERFERE may be used to subtract the self scattering (calculated by PLATOM) from the differential cross-section (produced by ANALYSE). The result is the distinct scattering, i(Q), (or 'interference term').

INTERFERE now works on GEM data. (as of 10-7-00)

INTERFERE is run from GENIE by typing @g_f:interfere (NOT from the a_b batch menu).

Renormalising

The simplest mode of operation for INTERFERE is just to do a subtraction of the self scattering for each detector group; i(Q)=I(Q)-IS(Q) (i.e. distinct equals total minus self). However, the result obtained may not oscillate about zero, as it should, and INTERFERE allows for such a discrepancy to be corrected in two different ways:

  1. The diffraction pattern, I(Q), for each group can be multiplied by a renormalising factor, alpha, prior to the subtraction. (This is my preferred method.)
  2. The diffraction pattern, I(Q), for each group can have a renormalising constant, beta, added to it prior to the subtraction.

The renormalising factors (or constants) for each group may be entered in two different ways:

  1. The renormalising factors may be typed in at the keyboard.
  2. The renormalising factors may be put in a file before running INTERFERE. This method is to be preferred because it reduces problems with typing mistakes, and it means that a permanent record is kept of the parameters used in the data analysis. Details of how to prepare the file are given here: factors, constants.

Caution should be exercised in using these methods to correct the data! If there is a marked and consistent discrepancy between the diffraction patterns and the calculated self scattering then it is probable that an error has been made at an earlier stage of the data analysis and this error should be corrected. For example, if the renormalising factors are markedly different from 1.0 then an error has probably been made.

After running INTERFERE it is possible to apply a further ad hoc correction whereby a straight line is subtracted from the interference function data for each detector group by running the LINT program. This allows the user to correct the data for a slight slope. Current experience with GEM data is that often such a correction is not needed.

Last updated on 23 Dec 2002 by Alex Hannon (a.c.hannon@rl.ac.uk)