Subarrays

[Barry Clark <bclark@aoc.nrao.edu>]

We seem to be agreed that we want some sort of scripting language for
the direct interface to the real-time system.  In designing this language,
a pivotal question is whether we want to give the observer the power to 
shuffle antennas from one subarray to another via script.  To do so complicates
things considerably, so I would like to close that door unless it is really
needed.

1.) It might be desired for transient phenomena.
One might want to record an instantaneous SED.  For phenomena with 
timescales greater than tens of seconds, it is more effcient to just 
cycle the array through the frequencies, rather than splitting into 
subarrays.  The faster timescale phenomena that come to mind are 
a). pulsars, b).  solar flares, and c). maybe flares on other stars
(the fine time-structure stuff may have such a steep spectrum that 
ALMA won't see it).  (Am I missing something?)

Note also that subarraying capability has to be provided for other reasons.
So, giving the observer the capability to rearrange his subarrays by 
script is called for only if this subarraying is pretty dynamic.  For
instance, the observer might want to occasionally recombine his subarrays
for calibration purposes (could use a weaker calibrator that way).

2.) It might be useful for instrumental reasons.
For instance, if the observer is switching between two frequencies, and
not all antennas are equipped with the receiver for one of these, 
he would like all antennas to join in when he is at the frequency that
they all have, but would have the incomplete antennas go off and do
something else possibly useful when he is observing at the other.

Am I missing other scenarios?

Do these two justify the complexity of including this in the scripting
language?
Date: Thu, 27 Jan 2000 11:40:49 +0100 (MET)
To: lucas@iram.fr
From: Majordomo@web3.hq.eso.org
Subject: Majordomo file: list 'alma-sw-ssr' file 'alma-sw-ssr.991118'

--