| Some months ago, Mark noted that when we observed very strong spectral lines
which spanned two chunks, the features seen in the chunk overlap region
sometimes seemed to have somewhat different frequencies in the two
chunks. Some chunk overlap regions showed the problem, others did not.
The magnitude of the error was not constant on different tracks. This
month, quite serendipitously, we happened to configure the correlator is
a very unusual configuration for testing purposes, which provided all the
information needed to make the problem appear to go away. It turns
out that all of our data, since day one, has had 1/2 of the chunks labelled
with frequencies and velocities 1/2 channel step too high, and the
other half have been labelled with frequencies and velocities 1/2
channel step too low. The sign of the shift depends on the sideband
used in the 2nd correlator mixing (the third mixing, if you count
the mixing at the SIS junction). So chunks s01 and s02 were shifted
one way, and s03 and s04 were shifted the other direction. This
meant that no problem was apparent if a line strattled the s03/s04
overlap region, as shown below:
but there was a one channel mismatch (1/2 from each chunk) in
the overlap regions such as s02/s03:
This problem has been fixed with , in an ad hoc way,
with a new version of dataCatcher, as
can be seen from this plot:
and this one:
Mark has worked out a sequence of MIR commands which will correct
the labelling problem in old data sets:
select,/p,/re,band=['s01','s02','s05','s06','s09','s10','s13','s14','s17','s18','s21','s22']
sp[psf].fsky = sp[psf].fsky + 5.e-4*sp[psf].fres
sp[psf].vel = sp[psf].vel + 5.e-1*sp[psf].vres
select,/p,/re,band=['s03','s04','s07','s08','s11','s12','s15','s16','s19','s20','s23','s24']
sp[psf].fsky = sp[psf].fsky - 5.e-4*sp[psf].fres
sp[psf].vel = sp[psf].vel - 5.e-1*sp[psf].vres
The new version of dataCatcher has been copied into the default area, so
data taken from now on should not need this correction. |
