[alma-config] Strawman array for costing

[Adrian Webster]

 A strawman array for initial tentative costing,        
 by Adrian Webster and John Conway, February 2000.
 
 In response to Mark Gordon's request for input of approximate array designs
for
 the purpose of making initial cost estimates, here is one possibility which
 combines the ideas of both of us. The design might be described as a 
 composite hybrid,  composite because it is made up of three
 configurations of different shape optimised to the constraints on
 different scales (i.e. dense pack, logarithmic spiral, and ring) 
 and hybrid because it is intended tozoom  smoothly all the way from the
 most compact to the most extended array. Details, such as the exact
 siting of the stations, are neglected and the required lengths of
 communications links are estimated from the general shapes of the
 configurations.
 

 1) The compact configurations. We suppose there may be two of these:
 
 a) a circular one, intended to give circular beams near the zenith. 64
 stations, as near together as possible, so on 15 m centres. The incremental
length of
 extensive services (road, power, fibre) is thus 15 m per station, for a
total of about 1 km of
 links.

 b) an elliptical one, elongated 2:1 north-south in order to give a
circular beam at a zenith angle of 60 degrees. 32 extra stations (it also
uses 32 of those in 1a above) typically 30 m apart, for another 1 km of
links.

 Total for both: 96 stations, 2 km of links.
 
2) The intermediate configuration.

A a logarithmic spiral similar to the one presented in Conway (2000) memo
283. The shortest route for the road and conduit links is not of course
along the 'conceptual spiral' shown in this memo but  instead
consists of roads along a small number (5 - 9) of curved radial 
'spokes' (see Figure 6, memo 260) linking the stations.
 
>From 15om to 1.5km radius on the array the total required
road/conduit length is about 16km, only weakly dependant on the 
exact number of spokes and branches used. If it turns out that  a 
circular  3km diameter 'A-array' is needed to give continuous resolution
between the <3km  spiral arrays and the 10km arrays as shown in Figure 2
of Memo 283 then an extra 4km of roads/conduits are needed. However with
careful  hybridization between the 10km ring and largesr 3km spiral this
extra
might be avoided. On the other hand some degree of interlinking between
spokes may be advantagous anyway for antenna transport and some 
contingency is needed to avoid difficult terrain, hence we include the
4km and give a conservative estimate of 20km of links.

The final estimated link length for the spiral portion of the array is 20km.

  The spiral design in memo 283 had 181 pads for all array sizes out
 to 3km. About 1/3 of these pads overlap with the estimate in 1 above 
 for dense pack circular and elliptical compact arrays. 

 Therefore a final estimate is 120 pads in the intermediate spiral
 portion  of the array.


 3) The outer configuration.
 
  A ring, in the form of a Reuleaux triangle probably or a circle possibly.
  64 stations. Links all around circumference, 10 pi km, roughly 30 km.
Radial
  links to outer part of the intermediate configuration, 3.5 km each, say 3
of them, 
 10 km in all. Total: 64 stations, 40 km of links.
 
 Summary:
 
  Compact           96 stations    2 km links
  Intermediate    120 sations   20 km links
  Outer                64 stations  40 km links
  
  Grand Total     280 stations  62km links 
  
 
 Adrian and John 

...........................................

Adrian Webster,
Institute for Astronomy,
Department of Physics and Astronomy,
University of Edinburgh,
Royal Observatory,
Blackford Hill,
Edinburgh EH9 3HJ,
Scotland.
asw at roe.ac.uk
tel: 0131-668-8391
fax: 0131-668-8416

...........................................