Alternatives to the Current CS Configuration

The CS configuration as currently conceived effectively removes two intermediate stations from the standard C configuration. Can we achieve the same gains in short-spacing coverage, by moving only one antenna? Or alternately, can we choose more wisely which two stations will be vacated, to minimize the holes created in the resulting uv-coverage? The options are restricted by the fact that at most 9 antennas may be placed on pads corresponding to a single arm. Figures - show the uv-coverage for six possible configurations, both with all antennas and with two (E14, W4) missing:

1.

C configuration: the old standard, shown here for comparison.

2.

$EW12\rightarrow EW3$: CS configuration as actually implemented in 1997.

3.

$EW10\rightarrow EW3$: we still move two antennas, but from the innermost stations which will not be occupied in D configuration. Since the VLA's uv-coverage is quite centrally concentrated, removing antennas from more central stations should result in somewhat smaller holes.

4.

$N10\rightarrow N1$: again we move an inner antenna, but this time only one, and to a more central location. The short-spacing coverage should be roughly the same as in Braun's (1993) original proposal (somewhat worse than 1997's CS configuration), but without sacrificing the resolution provided by N18. This configuration would require one additional antenna move, but this may be acceptable if the entire time allocated to C is indeed spent in CS.

5.

$W10\rightarrow W3$: as (3), but moving only one antenna.

6.

$W10\rightarrow W1$: as (5), but moving to a more central location; which would again require an additional antenna move from CS to D.

A few conclusions may be read directly from these plots. For short spacing coverage, moving two antennas is clearly better than moving one. From this point of view, there is not much to choose between taking antennas from EW12 or EW10. Removing EW10 (rather than EW12) does not give a significant improvement in the intermediate baselines: for long syntheses at some declinations it seems preferable to keep EW12 (e.g., $20^\circ$, $80^\circ$), at others EW10 appear more necessary (e.g., $40^\circ$, $60^\circ$). For snapshots the situation is even less obvious. Intermediate spacings on the other hand clearly improve when only one antenna is moved. From this point of view moving N10 to N1 seems marginally better than moving W10 to W3 or W1. The short spacing coverage is also somewhat better for the N1 option.