My understanding is that Andrey did not
apply any probe correction to this
data. The whole point was to keep it as simple as possible.
Attached is a summary of the results that I have. You will see that the
numbers that Andrey gets using his software are very similar to the ones
that I get by processing the far field data he sent me (which are I suppose
the same as he sent you). The values I get by processing the FF data you
sent me - in K-space - are significantly different. This is especially true
of the phase centre. It seems most likely that this is a result of the
interpolation which is done on the NSI data.
Note that the effect of the probe corrections is small, which is as it
should be because this data was taken with the low-gain "Andrey" horn.
Where we get into trouble is with the higher-gain horn. There the
correction is quite large and the results are quite a long way out, e.g. the
spill-over efficiencies of the set of scans 94 to 99, which were taken just
before this scan 103 but with the larger horn, are only
~91.5% when corrected with what we think are the right N's for that horn
and ~94% if no correction is applied. (But of course a correction should
be applied with that horn.)
- 26 Aug 2008