[daip] Possible major bug in ANTAB/APCAL

[Leonia Kogan]

Denise,

I need the data to solve the problem.
Can you ftp the fits file of the data you used as input for ANTAB?
You can ftp them to the aoc anonymous area:

ftp.aoc.nrao.edu
cd /pub/incoming/lkogan
put <the file>

Let me know please when it's done.
Send me the list of the parameters you used with ANTAB.

Thanks

Leonia

----- Begin Included Message -----

>From owner-daip at kochab.cv.nrao.edu Thu Jul 20 09:00 MDT 2000
Date: Thu, 20 Jul 2000 15:42:20 +0100 (WET DST)
From: Denise Gabuzda <gabuzda at jive.nl>
X-Sender: gabuzda at juw11
Reply-To: gabuzda at jive.nfra.nl
To: aipsmail at cv3.cv.nrao.edu
Subject: [daip] Possible major bug in ANTAB/APCAL
MIME-Version: 1.0
Sender: owner-daip at kochab.cv.nrao.edu
Precedence: bulk
X-Lines: 50
Status: RO


Some of our visitors have come across what looks like a rather 
strange problem.  They read in system temps and gains for an EVN 
experiment from an ANTAB-format file with a time offset of 90 
seconds to account for the fact that the Tsys measurements are 
actually just before the scans, rather than during them. These 
data appeared to be okay, but when they ran APCAL, the resulting 
solution table had only a few scattered entries. In other words, 
most of the dozens of system temperature measurements for each 
antenna were discarded by APCAL. We immediately thought of a 
problem with the specified time offset, but direct verification 
of the times in the TY table showed that it had been applied 
correctly, and that these times did fall within the times covered 
by scans. 

I've been working with a one-hour subset of that experiment, and
have verified the problem. In this case, although the times in
the TY table do fall within scans, the SN table made by APCAL is
completely empty. 

I initially got in touch with Athol Kemball about this, and he
directed me to this email address. His initial suggestions for
what might be wrong were:

i) inability to obtain a SOURCE_ID at the specified time from the 
index table; this should not be the case because running LISTR with
optype = SCAN yields a normal list of scan times and sources.

ii) an elevation cutoff (source is down); this is not the case,
during the times in question the source is definitely up.

iii) missing gain curve information; ANTAB writes that it reads
in all of the gain curves, and in addition if this were the
problem it is difficult to imagine how ANY of the input system
temperature points would be converted to gain values.

iv) opacity solution enabled in APCAL but incomplete source flux 
densities in SU table; no, the opacity solution was not enabled
in APCAL.

v) a bug; I'm afraid I'm having a hard time imagine what else it
could be!

I would appreciate it if someone could look at this. I can provide
a FITS file and ANTAB file for study/testing.

Thanks in advance,

	Denise Gabuzda
	Support Scientist, JIVE


----- End Included Message -----