[alma-config] Compact heterogeneous arrays

[Al Wootten]

Min Yun writes:
 > My worry exactly, but someone might insist on cross-correlating the
 > small dishes with larger ones -- I can argue both ways.  Exactly
 > what the ultracompact array is and what its intent and operation needs
 > are to be better defined.  For example, should they be reconfigurable?
 > If not, we could just stick them in the middle of the current 
 > configuration designs, perhaps on a higher pedestal as you suggeste.
 > If it were to operate as an independent high frequency array, then
 > it would make sense to place it physically apart from the main array.
Stephane presented a plan at the ALG meeting.  I think Peter has a copy
but I do not.  I asked Peter to see if he did, and send me a copy.
SG showed some considerations; I will type from my notes.  
Morita also presented a plan.

Guilloteau:
To increase the parameter space for science, one option is an array of
small antennas, to fill the short spacing gap, which occurs at spacings of
about 6-10m.  A compact array is sufficient for this problem, and
7 8m antennas is thought to be enough if mosaicing occupies 25% of the
time.

The goals of this array are to 
 a) Directly measure 8m spacings
 b) Improve high frequency performance at 900 GHz and perhaps above.
 c) It may help to cross calibrate the total power/interferometric modes on the
     array

Note that it may be possible to make a smaller antenna with a much better
surface than the 12m.  If, perhaps, a 10 micron surface were possible the
collecting area available may complement, at some high enough frequency,
the rest of the array very nicely.  It may be possible to underilluminate
the 12m somewhat at the high frequencies, with these smaller antennas
available to fill in the close spacings, relaxing pointing and field of
view restrictions on the large antennas.  One might employ longer baselines
using the compact (fixed) antennas and underilluminated 12m antennas.  There
would probably be no need to cross correlate with the larger antennas at
short spacings or for low angular resolution.  One would cross correlate for
higher angular resolution.  One might achieve performance above 1 THz.  

Morita showed a small antenna compact array.  He showed some studies
by Viallefond I believe on 15 x 6m antennas, including a central cluster, 
five 3 antenna clusters, and 3 5 antenna clusters.  My notes say that
the last was the best match in weights.  Stephane noted that if the
antennas could be closely enough packed, a larger (8m) diameter would
be good, but there was some question as to how closely one could pack
them and move them quickly without colliding.  A simulation study is needed
to determine the ideal diameter and number.  It was noted that if they were
on the same mount, same receiver cabin, there would probably be a cost
savings.  It is not clear how that affects packing, however.
 > 
 > I would like to raise this issue at the
 > ASAC meeting this week, and I would like to ask Morita-san to
 > give a presentation at the Tucson meeting concerning this
 > concept.  If Morita-san can give us an early preview as what the
 > current concept is, that might help us to work out some of the
 > details in advance.
That would be useful.

 > p.s. If such an array is part of ALMA, surely this will require a
 > subarray capability of its own at all times, and this means the
 > standard operation mode of ALMA would be at least 3 subarrays at
 > all times, with 4 or 5 subarray operation quite often.  This has
 > a major implication to the to the system design, and this needs
 > to be communicated clear to the rest of the project!
Well spoken.  I put Steve's subarray note on the ASAC reading list,
which makes this point at the end.

Al