[daip] forwarded message from Chris Phillips

[Eric Greisen]

Is he likely to be right?  I suspect so but thought I had better ask.

E
------- start of forwarded message (RFC 934 encapsulation) -------
Received: from polaris.cv.nrao.edu (polaris.cv.nrao.edu [192.33.115.101])
	by zia.aoc.NRAO.EDU (8.9.3/8.9.3) with ESMTP id JAA24269
	for <egreisen at aoc.nrao.edu>; Wed, 6 Mar 2002 09:39:45 -0700 (MST)
Received: from cv3.cv.nrao.edu (cv3.cv.nrao.edu [192.33.115.2])
	by polaris.cv.nrao.edu (8.9.3/8.9.3/CV-SOL-3.0) with ESMTP id LAA15059
	for <egreisen at polaris.cv.nrao.edu>; Wed, 6 Mar 2002 11:39:44 -0500 (EST)
Received: from server4.nfra.nl (fwuser at firebox.nfra.nl [192.87.1.200])
	by cv3.cv.nrao.edu (8.9.3/8.9.3/CV-CV3-4.0) with ESMTP id LAA08029
	for <egreisen at cv3.cv.nrao.edu>; Wed, 6 Mar 2002 11:39:43 -0500
Received: from juw17.nfra.nl
	by server4.nfra.nl; Wed, 06 Mar 2002 17:40:25 +0100
X-Sender:  <phillips at juw17.nfra.nl>
Message-ID: <Pine.GSO.4.30.0203061707410.6571-100000 at juw17.nfra.nl>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
From: Chris Phillips <phillips at jive.nl>
To: Eric Greisen <egreisen at cv3.cv.nrao.edu>
cc: Huib Jan van Langevelde <langevelde at jive.nl>
Subject: CVEL for JIVE data
Date: Wed, 6 Mar 2002 17:39:21 +0100 (MET)

Hi Eric,

One year later, I have finally had the time and the right experiment
(1000 Jy methanol maser emission from W51) to test whether CVEL works
properly with JIVE data.

In a Nutshell, it doesn't. Basically the MarkIV correlators do fringe
rotation on a baseline basis (like the MarkIII correlators), but the first
antenna is rotated for MarkIV not the second antenna. For JIVE data, aips
assumes the first antenna was the reference antenna and gets the frequency
shift wrong. For EVN baselines with is of the order of 0.5 km/s error.
(For global baselines it is more like 2 km/s).

I got it to work with a change in DSHIFT. Basically I made the following
change:

C      IF (IREFST.EQ.0) ITEL = IANT1
C      IF (IREFST.GT.0) ITEL = IANT2
      ITEL = IANT2

So, is it possible to either:

 - Put a test in CVEL for either JIVE or a MarkIV correlator and set ITEL
   to the correct value (ie IANT2)
 - The IREFST set above is obviously meant to deal with this problem.
   However from memory you didn't think it was implemented in the rest of
   aips so it wouldn't work.


Let me know if you want any more info from me (or more justification for
my request, but I guess you don't really want to see a million plots
which show that CVEL works best with the above fix).

Cheers
Chris Phillips



------- end -------