[daip] problem with GC table loaded with FITLD

[Amy Mioduszewski]

Hi Eric,

Adam Deller reported a problem with the GC table loaded with FITLD where 
the second IF is missing.  (see dicussion below) I loaded the data and 
confirmed the problem which exists in 31DEC15/16 but not 31DEC14.  I 
looked at data I loaded in November and it is fine, loaded that same 
data again and all the IFs but the first were missing, so this is a 
recent problem.  The symptoms of the problem are that there is an entry 
in the GAIN column for the first IF and then INDE thereafter, the SENS 
column seems to be filled in properly.  Do you think your change of Jan 
7 might have done this? If you want to test this you can copy Adam's 
data from /home/thor2/amy/bb351b.adam.fits, the dataset is relatively small.

Cheers,

Amy
-------- Forwarded Message --------
Subject: 	Re: bb351b is correlated
Date: 	Thu, 14 Jan 2016 21:47:19 +0100
From: 	Adam Deller <deller at astron.nl>
To: 	Amy Mioduszewski <amiodusz at nrao.edu>
CC: 	VLBI Observers <vlbiobs at aoc.nrao.edu>



Hi Amy,

Thanks for looking in to this.  I can FITLD the data with 31DEC14 as a 
work-around.

Cheers,
Adam

On Thu, Jan 14, 2016 at 8:42 PM, Amy Mioduszewski <amiodusz at nrao.edu 
<mailto:amiodusz at nrao.edu>> wrote:

    Hi Adam,

    I have loaded the data with DEC14, DEC15 and DEC16 and for DEC15 and
    16 the second IF's gain is missing (INDE).  I looked at some data
    that I loaded in in November and it is fine, I loaded that data
    again and it has this problem, so it is something recent.  There is
    a note to a change to the GC table in CHANGES.DOC Jan 7 that was
    made to both 2015 and 2016.  I will report this to Eric, since he
    made the change, but he doesn't seem to be here today so it might
    have to wait until tomorrow.

    Amy


    On 01/14/2016 10:33 AM, Adam Deller wrote:
>     Hi Paul, Amy,
>
>     I confirm that there is actually nothing wrong with the GAINS
>     tables in the fits-idi file.  But when I use 31DEC15 AIPS to load
>     it up with FITLD, I lose the entries for the second IF.  Even
>     stranger though - 31DEC14 works fine.
>
>     So there is something weird going on with FITLD.  Could one of you
>     check in Socorro whether you see this behaviour with local AIPS
>     31DEC15?  (I have been testing the file
>     VLBA_BB351B_calibb_BIN0_SRC0_1_160114T160128.idifits).
>
>     I can also try installing 31DEC16 to see if it is the same as 15.
>
>     Cheers,
>     Adam
>
>     On Thu, Jan 14, 2016 at 5:47 PM, VLBI Observers
>     <vlbiobs at aoc.nrao.edu <mailto:vlbiobs at aoc.nrao.edu>> wrote:
>
>         Hi Adam,
>         Very strange. I spoke to Amy and there isn't anything
>         obviously wrong about the gains file. She suggested I remake
>         the fits files and have you look at the new ones. I've done
>         this (the widefield is still in progress) Can you get the new
>         files and see if they are okay?
>
>         Cheers,
>
>         Paul Dyer
>         NRAO Data Analyst
>         pdyer at nrao.edu <mailto:pdyer at nrao.edu>
>
>         On 01/14/2016 04:58 AM, Adam Deller wrote:
>
>             Hi Paul,
>
>             The GC table for bb351b is busted - the second IF has no
>             good values.  That
>             leads to crazy values in APCA for the 2nd IF, and
>             ultimately an amplitude
>             scale that is out by 10 orders of magnitude :)
>
>             Probably some problem with the data provided to difx2fits
>             - can you check,
>             please?
>
>             Cheers,
>             Adam
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listmgr.nrao.edu/pipermail/daip/attachments/20160114/d76167df/attachment.html>