[evlatests] Large Bz error?

[Vivek Dhawan]

Looking under the lamp post for the usual things:

EOP looks correct - I compared the numbers in the
WIDAR data header to the USNO file here:

http://gemini.gsfc.nasa.gov/solve_save/usno_finals.erp

I next tried to compare the XYZ antenna positions in the
AIPS header (via SDM and CASA) to the VLA pad positions
in the VLBA database which purport to be in ITRF:

$SCHED/catalogs/locations.dat

There is 20-30cm of difference in all Y & Z axes,
between a pad's location in the WIDAR header and its
location in the SCHED/USNO database. Its not just
a longitude shift since Z is involved. The error is
(to within ~1cm) independent of which pad I compare.

It is also not clear whether including the continental
drift term (12.73 years since the USNO solution epoch)
is useful or not - X is better with the correction, but
the discrepancy is about 30cm in at least 2 axes with
or without the drift put in.

Where do those WIDAR positions come from?


-------- Original Message --------
Subject: Re: [evlatests] Large Bz error?
Date: Sun, 27 Sep 2009 14:33:59 -0600
From: Vivek Dhawan <vdhawan at aoc.nrao.edu>
CC: evlatests at aoc.nrao.edu

WIDAR data from Sept 01 also show this - but I think its more
of a time error.

The phases after transfer from 3C286 to the 2 other targets show
residuals upto of 100deg, of opposite sign on the farthest E and W
antennas. The two target sources are Both about 1.5 deg away from
3C286 in declination, in the same sense, but their RA offsets are
<0.5 and ~4deg, and the farther one shows a much bigger EW phase diff.

I think this is more like a time error of ~0.5 second(!)
(I get this as follows: ~90deg @ 6cm =  1.5cm error over 1/10 radian
throw or 15cm at ends of C array, (Sep 1 was still C array)

This is only on WIDAR data, the VLA baselines data from sep 23 and 26
show only ~1cm offsets from the antenna moves, and with no systematic
dependence on arm position as would result from time offset.


Rick Perley wrote:
>     Further observations based on the Thursday night data ...
> 

> 
>     2) There is a very clear, and very large, time independent offset in 
> phase between the four sources.  The offset is clearly a function of the 
> source declination.  Such an offset is most likely due to a error in the 
> antenna position z-coordinate.    Not all antennas show this effect, but 
> those that show it strongly are all at the ends of the arms. 

snip

>     These position errors should explain why phase calibration of an 
> object which is a few degrees away from a calibrator returns a very poor 
> image. 
>    
> _______________________________________________
> evlatests mailing list
> evlatests at listmgr.cv.nrao.edu
> http://listmgr.cv.nrao.edu/mailman/listinfo/evlatests
>