Recommendations of Design Review Committee on ALMA Configuration Design Review Socorro, New Mexico, 24-25 January 2002 Al Wootten, rev 1.1 Committee Members: ALMA Division Heads/IPT Leaders Peter Gray (pgray@nrao.edu), Stephane Guilloteau (guillote@iram.fr), Simon Radford (sradford@nrao.edu), Gie Han Tan (ghtan@eso.org), Al Wootten (awootten@NRAO.EDU External Reviewers: Douglas Bock (dbock@astron.berkeley.edu), David Woody (dwoody@ovro.caltech.edu), Mel Wright (wright@astro.berkeley.edu) The committee appreciates the efforts which configuration designers put into the designs which were presented for review at the meeting. Several novel concepts merit special notice in the "General Comments" section below. General comments 1) The design of configurations for radio telescope arrays has matured, the committee commends the number of new ideas introduced during the course of ALMA design: the elegant design theory developed by Boone, the sidelobe optimization techniques presented by Kogan, the Delaunay uv optimization introduced by Webster, and the hybrid uv then beam optimization employed by Conway. 2) High zenith angle observations present particular problems in the most compact configuration, when shadowing may interfere. Solving this problem without exceeding the antenna pad limits is challenging; the committee found Webster's analysis and solution to this challenge worthy. 3)Webster's zoom array designs incorporated the idea of the array zooming out in more circular configurations suitable for zenith observations, then zooming back in more elliptical configurations. This idea may have operational advantages during poorer weather seasons when skies are more transparent closer to the zenith than later during the zoom back to more compact configurations at a later more transparent season. 4) The committee endorses the recommendation of the Preliminary Design Review that the number of antenna pads be kept below 250 and notes that none of the configurations presented exceeded this limit (or: notes that several designs employed substantially fewer?) 5) The configuration design effort has fallen behind earlier schedules and requires substantial effort to remain within the current ALMA metaschedule. Specific details: Given point 5) above, the committee feels the designs presented form a good basis for improvement and development into the final design methodology for ALMA configurations. Detailed optimization of the configuration layout, synergistically with road, cable and fiber connections to antenna pads, and final transporter design will be considered at a later Antenna Configuration Engineering design review. To effect this we recommend: 1) The compact array poses especially difficult design challenges and will be the first to be constructed; it is the kernel from which the subsequent configurations proceed. The Kogan designs for the compact array have been communicated to the antenna transporter group and incorporate feedback from them. The detailed plan for antenna extraction and movement to the next larger configurations looks promising. However, the committee believes that the highest demand for mosaicing will occur for the compact array, and is concerned that the distant sidelobe structure may not be ideal for these observations. The beams delivered by the Boone optimization appear to have especially good characteristics. The committee recommends that a mask approximating the Kogan design be used in conjunction with the Boone code to improve the beam. Attention needs to be devoted to the plan for observations of high zenith angle sources with this array, and to finding the best compromise between high brightness sensitivity and shadowing. The design process should be followed by imaging simulations to verify excellent mosaic performance. Some iteration may be required during this procedure, and also with the tranporter design group. Work needs to be devoted to finishing this design by April 2002. 2) The compact configuration should then be mated to the next larger configurations. The committee found the Conway design most complete from the point of view of scientific capability and operational planning among those presented for the intermediate configurations. It should be optimized using the Boone approach but preserving the large spatial dynamic range demonstrated in the Conway design. Some short spacings need to be incorporated in larger arrays to achieve this. Simulations can more straightforwardly suggest when too few short spacings are present than when too many are present. The committee therefore recommends three or more antennas providing short spacings be included with simulations guiding the final choice of a sufficient number. Shadowing performance must be minimized for these three antennas for a wide range of elevaions. The largest intermediate configuration should be of a dimension 4.5-5 km. 3) The largest configuration poses challenges of a different sort. A 14km configuration should be optimized without undue expense of a hybrid to the 5km configuration. Elevation differences should be taken into account in the optimization but should not drive it. The 14km array should demonstrate reasonable snapshot performance. Use of the road to the Operations Support Facility should be studied, taking into account implications for the system design (correlator delay run length). 4) Along with these designs a detailed sequencing plan for moves between configurations will be presented in a manner which is clear for comment and input by ALMA Operations planning.. The committee expects quasi-continuous antenna movement to be employed during operations, except between the extreme arrays, when antenna movement will need to be in a burst mode. Actions Needed 1) The project should appoint a leader of the configuration design process as quickly as possible. The leader will guide the design process, including iteration of the design with the antenna/transport groups and the site development group. The project will invest in this leader the resources necessary to be able to push the process ahead as a top priority without distraction from other commitments. 2) The project should contract with a consulting engineer as soon as possible to provide a first estimate of costs involved in construction of the configurations on the site, and to provide initial input on road and utility conduit design to the configuration leader for iteration to a solid plan for ALMA construction. The project should invest this engineer with the resources necessary to finish the configuration design process with high priority. The transporter design interacts with the configuration design and optimization; a strong engineering effort from the project on transporter design is needed to meet the configuration deadlines. 3) The project should proceed with obtaining high resolution imagery and maps, particularly of the central regions of the site where most of the antenna pads will be located. 4) This report and an update of progress through the second step above should be presented by 'ALMA Week' at the end of April. The status of configuration design site planning integration will be reviewed during the Site Development Preliminary Design Review planned for September 2002. The final configuration design process will be complete by January 2003. The final configuration design will include deliverables itemized below. Deliverables 1. Projected antenna locations need to be visually inspected on aerial photographs, and problem areas incorporated into the mask. The project should contract for a survey of sufficient detail for engineering studies. 2. Compact configuration design, including high zenith angle versions a. Simulations, using images from the Image Library, i. snapshots and ii. long tracks, iii. Mosaics iv. from declinations from -70 to +45 degrees. b. Shadowing studies of the compact configuration c. Discussion of the interaction between compact array observations and observations with dedicated single antennas (i.e. how many are needed to equalize brightness sensitivity over how much time) d. Iterated with Transporter Design Group e. Report during ALMA Week 23-27 April 2002 f. Iterated between consulting engineer and configuration leader to produce plans for roads, cables and fiber by September 2002 Site PDR 3. Intermediate configuration design, including high zenith angle versions g. Simulations, using images from the Image Library, for i. Snapshots ii. Long tracks iii. Mosaics iv. Declinations from -70 to +45 degrees h. Large spatial scale dynamic range (>50 as a target), to include short spacings from three antennas in the above simulations i. Shadowing studies j. Largest baselines 4.5-5 km k. Iterated with Transporter Design Group l. Iterated between configuration leader and consulting engineer to produce plans for roads, cables and conduits by September 2002 Site PDR 4. Largest configuration with baselines around 14km m. Simulations, using images from the Image Library, for i. Snapshots ii. Long tracks iii. Declinations from -70 to +45 degrees n. Elevation differences should be taken into account o. Tradeoffs between configurations ringing Chascon and Y-type configurations employing the OSF road should be discussed. 5. A clear plan for moves between configurations should be presented p. Iterated with ALMA Operations planning q. Iterated with Transporter Design group