[mmaimcal]Re: first step on compact config design evaluation

[Al Wootten]

Hi Mark

I have circulated a draft of the committee recommendations for iteration but
I don't want to circulate a draft here before it reaches convergence or at
least one iteration, for fear of too much confusion.  
But my opinion of what the committee wants is that we 
concentrate on the Kogan design at this point as the others
are less well-developed.  Of the several designs Leonia devised, the 2 road
design John used perhaps supplemented by the one road design should be the
focus of attention.  What the committee preferred to see was that
the tests you propose be applied to the Kogan array.  The committee felt
that mosaicing quality might be better if that array were refined with a
Boone-type algorithm which might result in a better beam.  I see Leonia's
message of this morning which suggests that the committee's intuition may
have been misguided; simulations will surely help rectify that.  

Woody's beam calculations are available at:
http://www.ovro.caltech.edu/ovrodocs/ALMA/ALMA_configs_Boone_20JAN02.pdf
http://www.ovro.caltech.edu/ovrodocs/ALMA/ALMA_configs_Conway_20JAN02.pdf
http://www.ovro.caltech.edu/ovrodocs/ALMA/ALMA_configs_Webster_20JAN02.pdf

In any event, the committee felt that uv optimization might be beneficial
to the Kogan design.  It wasn't sure that a further beam opimization,
as John applied to some of his designs, was critical.  I discussed this
with John some more yesterday and he agreed that the last beam
optimization step didn't result in very large changes in his work.
The problem may come with the uv optimization--it isn't clear that
Boone's code can be effectively used or changed by others.

John will lead the configuration effort but he went to Bonn for a 
mm vlbi meeting so let's discuss this tomorrow at the meeting then
send the plan to John for his input.

Clear skies,
Al
Mark Holdaway writes:
 > 
 > I am aiming to evaluate the compact configuration designs
 > and to eventually help shape the 2 or 3 compact configurations.
 > This is my suggested starting point, and I am putting this out to
 > you for comments.
 > 
 > What to evaluate:
 > 
 > 	* mosaic image quality
 > 	* sensitivity losses due to shadowing
 > 	* beam elipticity
 > 
 > This order reflects my opinion of the importance of
 > each item.  However, the sensitivity loss and beam elipticity
 > will determine if we should even bother looking at the image quality.
 > 
 > As a starting point, I propose to simulate mosaic images
 > with a) thermal noise and b) pointing errors, 
 > for declinations
 > 
 > -70 -60 -50 -40 -30 -20 -10 0 10 20 30 
 > 
 > The full declination range
 > -90 ==> +60 will be simulated later as we investigate the
 > extreme delcination configurations.
 > 
 > 
 > We have three configurations:
 > 
 > Webster: the most compact (probably not possible)
 > Kogan: intermediate
 > Boone: the least compact
 > 
 > ==============================================================
 > 
 > Ultimately we want a complete set of configurations for all
 > declination ranges.  At the very least, this means selecting
 > a compact configuration and then making one or two hybrids from
 > inner Conway intermediate array stations.  At the other end of
 > this process, we could drive the inner parts of the Conway array.
 > 
 > I do not see a closed loop in which simulation results drive the
 > relocation of stations heuristically.  In fact, I am not sure how
 > to drive the relocation of stations.  Any ideas?
 > 
 > 	-Mark
 > 
 > 
 >