[mmaimcal] Optimizing arrays for LMA

[Tamara Helfer]

Notes on LMA configuration design	-for discussion at MMAImcal
---------------------------------	meeting today (99feb01)

We assume 48 antennas of 12m each, with a minimum separation of
1.25 D or 15m.

If we again take a filling fraction of .4 for D array, then
the (circular) D array diameter is about 130m, up from 95m for 
the 36x10m case.  If we fix A array at 3 km still, then we get 
a resolution scale factor of about 3.2 instead of 3.6, a 10% 
reduction.

As a starting point, I've chosen  C, B, and A array minimum and 
maximum annuli for donut configurations based on some guesses as 
to what are reasonable compromises between sidelobes and uniformity.
Based on Adrian Webster's and John Conway's memos, the Europeans may 
not sign on for fixed arrays for the intermediate cases, but I think 
this is still a reasonable starting point for the combined array.

(1) D array: Filled array with d=130m. 

(2) C array: Donut array with dmax/dmin=0.5; dmax=366m, dmin=183m.

(3) B array: Donut array with dmax/dmin=.75; dmax=1040m, dmin=781m.

(4) A array: Donut array with dmax/dmin-0.9; dmax=3000m, dmin=2710m.


Leonia is working on producing sidelobe-optimized arrays given these 
constraints; he has already produced an optimized C array.