[daip] APCAL issues.

[Craig Walker]

I have been doing opacity corrections with APCAL and having some problems.

At the moment, I am trying with solint=480 (=8hr) on a data set that is
16 hours long.  I get three solutions per station and the plots and
times suggest that the middle one only used a very short segment of
data.  I tried lengthening the solint and still get 3 solutions.
I then tried shortening the interval to 300 sec.  From the plots, it
looks like what is happening is that you get a 300 second solution, then
a solution with about 2 points in it, then another 300 second solution,
then another very short one ....  This should be the clue needed to 
determine the problem.

Does Trec need to be R,L pairs even if you only have one polarization?
(I have both).  Are the elements of Trec based on antenna table antenna
number, or on elements in ANTENNAS?  If the former (as I guess from the
help file) the program cannot be used for more than 15 station
observations.  That's a pretty bad limitation.

The plots go to very high sec(z) - like 300 in some cases.  This is
not useful and makes it hard to interpret the interesting range which
is all scrunched up in the left end.  I suggest either a reasonable
max sec(z) be chosen, or the be APARM(5), or some such.


A general aips issue that comes up if you want to specify TREC for APCAL:
The limitation to 80 columns when giving a long array can be a big pain.
You end up having to give the array elements one at a time.  Is there
any hope that the parser could be extended to more than 80 columns?  We
aren't using card punches any more.  The other possibility would be to
have some way to give a list of elements not starting at the first.