Fwd: Re: [daip] APCAL

[Eric Greisen]

I just took a look at your problem and AIPS does not handle it
correctly at present - I am working on it.

When you make N SN tables, one for each IF with APCAL and then run
CLCAL with MERG you end up with the N tables concatenated (and time
sorted), but not really merged.  There is sort of a work-around, in
that you can time smooth with a very short boxcar and DOBLANK=1 and
this will replace the magic blanks written by APCAL (for all IFs
except the one being done in each SN table) with the good values from
the other tables.  But there is an error in smoothing the delays in
that magic blanks are not caught and so phases are adjusted because it
things delays were.

I am in the midst of fixing these things - first to fix the smoothing
for the blanks and then to do a real merge operation in the merged SN
table - as Amy seems to assume is already done.

The 31DEC05 version in a few days (or maybe tomorrow) should have
these corrections.

Eric Greisen