[evlatests] last bit of VLA data still bad

[Walter Brisken]

Hi Michael,

It is clear the obs2script bug fix is in and working.

I claim that the last source of your file ended at 18:04:20 IAT and that 
everything IDCAF did was fine.  after 18:04:20 IAT, a source called FINISH 
at 18:00:00 RA 00:00:00 dec was configured.  This is the standard source 
that informs IDCAF not to write more records.  This source was observed, 
but not recorded, until 18:05:06 IAT.  It seems the odd end effects you 
are seeing are something completely different from previous end-of-expt 
problems.  It is suspicious that it is the last 10 seconds of a source and 
that a second subarray was running at the same time (sysstartU).  I don't 
think this is an IDCAF or obs2script problem.  If you see a source named 
"FINISH" in your database then something is wrong with idcaf.  I'm not 
sure what could cause screwy amplitudes in the last 10 sec of the file. 
It would be good to get feedback from others doing tests or observations 
with single subarrays.

WANTED FROM TESTERS -- reports on the end of tests or observations -- are 
the last unflagged records good or not?  Does the observation end at the 
right IAT time?

-Walter

On Mon, 7 Jan 2008, Michael Rupen wrote:

> Hi Walter,
>
>  this was a dynamic file, AR642-A-B-O (VLAOBS 'AR642'), observed today
> (7jan08= MJD 54472). According to the operator's log, the observations ran
> from 17:04:00 to 18:04:00 IAT.  This ends with an X-band scan on 1407+284 = 
> OQ208.  On-line FILLM with on-line flags turned ON produces a file with this
> last scan running from 18:03:12 to 18:04:18.  The last three 3.3s
> integrations of this scan show odd (scattered) amplitudes.  I'm guessing this
> is a flagging problem, with the file ending but data continuing to be taken
> during the transition to the next file.  But you'd know better than I!
>
>  Cheers,
>
>            Michael
>
> p.s. Bryan has not been in Socorro today, so don't be surprised if he doesn't
>  respond quite as instantaneously as usual :)
>
>
> On Mon, 7 Jan 2008, Walter Brisken wrote:
>
>> 
>> Bryan -- has the obs2script fix I put in a bit ago been built?
>> 
>> Michael -- can you send me the essentials -- obs file, project code, mjd, 
>> UTC of file end, and what you saw after the purported end of file?
>> 
>> -W
>> 
>> On Mon, 7 Jan 2008, Michael Rupen wrote:
>> 
>>> 
>>> 
>>> ---------- Forwarded message ----------
>>> Date: Mon, 07 Jan 2008 13:58:23 -0700
>>> From: Rick Perley <rperley at aoc.nrao.edu>
>>> To: Michael Rupen <mrupen at aoc.nrao.edu>
>>> Subject: Re: last bit of VLA data still bad
>>>
>>>   Yup -- Walter claims it's fixed, to which Barry says 'says here'.
>>>   But clearly it's not.  You'd best inform the two of them.  I never know 
>>> which of them is in charge of what ...
>>>
>>>   Rick
>>> 
>>> Michael Rupen wrote:
>>>> Hi Rick,
>>>>
>>>>   I'm just looking at some data taken a couple hours ago -- the last 10s
>>>> of the last scan are rubbish (fringes but unrelated to the rest of that 
>>>> scan).
>>>> Am I right in thinking I heard you say Walter put in some fix that should 
>>>> have eliminated this sort of problem?  If so, I'll forward to him.
>>>>
>>>>   Thanks --
>>>>
>>>>            Michael
>>> 
>> 
>