[daip] Re: Help please

[Leonia Kogan]

Cris,

Do you want to comment?

Leonia

-----------------------------------
Johan,

I am glad that you could finally handle GETJY

>I see that if I followed Appendix B of the Cookbook I would have seen
>that one has to use all the calibrators. I think it is somewhat unclear
>in Chapter 4 though.

I agree with you. We are working here improving this part of the cookbook.

>So I did that and ran getjy afterwards and indeed flux densities were
>assigned to each source. That's good. But there is still one problem: I
>have quite a number of calibrators and typed in all their names.
>However, when doing "inp vlacalib" only the first 7 or so appeared and
>the last couple of calsources seemed to be ignored and also seemed not
>to have been included in the calibration.

Typing INP AIPS does not reproduce precisely the inputs. 
For example AIPS prints the reals with less number of digits for real inputs.
The number of printed CALSOURs is limited also.

But it is not a problem. I am pretty sure that VLACALIB finds solution 
for all listed calibrators if the number of calibrators does not exceed 30.

>Closure errors at 0/22:07:50 1743-038   IF no 1 Rpol
>Average closure error  2.368%   1.23d

This means that closure error at the given time is ~2% by amplitude 
and ~1degree by phase 

>03-07 36.4% 7d 03-23 10.7% 1d 07-20 28.2% 18d.

This means that baseline 3-7 gives deviation from the solution ~36% 
by amplitude.

To edit or not to edit. This is the question. It is rather subjective.
My opinion is that the errors are not so big so you do not need editing.
But ....
and 7 degrees by phase ....


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

>From johan at fskdjvdw.puk.ac.za Thu Mar 30 02:44 MST 2000
Date: Thu, 30 Mar 2000 11:44:48 +0200 (SAST)
From: Johan van der Walt <johan at fskdjvdw.puk.ac.za>
To: "[Leonia Kogan]" <lkogan at zia.aoc.NRAO.EDU>
Subject: Re: Help please
In-Reply-To: <200003291519.IAA02089 at bonito.aoc.nrao.edu>
MIME-Version: 1.0
Content-Type: MULTIPART/MIXED; BOUNDARY="-1601233368-474257721-954409488=:5109"
Content-Length: 284497
X-Lines: 4684
Status: RO

Leonia,

I see that if I followed Appendix B of the Cookbook I would have seen
that one has to use all the calibrators. I think it is somewhat unclear
in Chapter 4 though.

So I did that and ran getjy afterwards and indeed flux densities were
assigned to each source. That's good. But there is still one problem: I
have quite a number of calibrators and typed in all their names.
However, when doing "inp vlacalib" only the first 7 or so appeared and
the last couple of calsources seemed to be ignored and also seemed not
to have been included in the calibration. Do I have to run vlacalib
more than once? What about the SN table? Generate more than one and use
all in GETJY? 

I'm attaching the output of vlacalib. It is quite a lot of paper so
don't print if necessary. There now are more solutions obviously. 2408
good solutions, failed on 94. If it falls within the line of your
duties as the assigned AIP, is it perhaps possible to please look at it
and make comments on what editing I still have to do. Eg. On the first
page of the output which lists the errors etc. for calibrator 1743-038
there are a couple of lines saying


Closure errors at 0/22:07:50 1743-038   IF no 1 Rpol
Average closure error  2.368%   1.23d
03-07 36.4% 7d 03-23 10.7% 1d 07-20 28.2% 18d.

Are these specific baselines at  0/22:07:50 that have to be flagged? 
For 03-07 and 07-20 the closure errors are significantly larger than
10%. I used TVFLG to check on eg baseline 03-07 and flagged it but it
had negligible effect on the closure error for that baseline when I ran
vlacalib again.

Regards

Johan




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