Low-Frequency Aperture Array


The Aperture Array Design and Construction Consortium had a productive period with successful milestone achievements and a good ramping up to the production and roll-out of the Aperture Array Verification System (AAVS1), a 400-antenna 4-station system, which will be the ultimate LFAA test platform.

LFAA Antenna Progress: a new Antenna Design

We have now realised the design of SKALA-3. With minor modifications in the input network of the LNA and the size of the bottom dipole of the antenna we have improved the passband smoothness of SKALA by removing sharp spectral features that could potentially impose a constraint in the calibration of the passband for the 21-cm cosmology experiments (Cosmic Dawn and Epoch of Reionization). We have measured the resulting design using local low-order polynomial fits over different bandwidths as well as against the metric used for the Delay Spectrum technique used by the precursor HERA. More information can be found in de Lera Acedo et al. 2017 (submitted to MNRAS, https://arxiv.org/abs/1702.05126)) and in Trott et al. 2017 (submitted to MNRAS). SKALA-1, -2 and -3 have been specifically designed to meet the SKA L1 requirements across the full frequency band: 50-350 MHz, including sensitivity, cross-polarisation, etc. We can, however, compare its passband smoothness to that of antennas designed for other EoR experiments/telescopes with different capabilities, bandwidths, design requirements/strategies and metrics e.g. HERA and the MWA dipole. This is done in the 2 aforementioned papers showing similar or better performance of the SKALA-3 antenna in the bandwidths used for the comparison. SKALA-3 also fulfilled the requirements derived by Trott in a 2016 paper (https://arxiv.org/abs/1604.03273).

Picture1Figure 1. First prototype of SKALA-3 at Lords Bridge, Cambridge, UK (left and centre). Ready for testing. On the right, the arms of SKALA-3 and SKALA-2 compared.

The Aperture Array Verification System, AAVS1

Production SKALA2

The plastic parts of the pyramids were fabricated using injection moulding techniques. The metal parts were fabricated using low-cost wire bending techniques and they were coated for protection against the weather in Western Australia. The LNAs were fabricated and tested by a local company (see Figure 2). Furthermore, the pyramids (pre-assembled with the RFoF modules and Hybrid cables in Italy) were assembled (see Figure 3.) and tested by the same local company and were shipped direct to Western Australia.

Picture2Figure 2. AAVS1 LNAs being tested by an engineer from a local company.

Picture3Figure 3. The trumpets were labelled appropriately to ease the assembly and deployment process.

Roll out first batch AAVS1

During a 10-day installation campaign from 13 to 22 March, an international team of engineers from Australia, Italy, Malta, the Netherlands, and the United Kingdom installed 96 prototype SKALA2 antennas, along with the Tile Processing Modules (TPMs) and related firm and software at the Murchison Radio-astronomy Observatory in outback Western Australia, the hosting site of the Murchison Widefield Array, MWA.

Apart from a few ongoing engineering questions, the site-trip was considered a success and many lessons were learned towards the next AAVS1-deployment, as well as towards SKA1-low.

Picture4Figure 4. AAVS1 opening ceremony.

Picture5Figure 5. Group picture of AAVS1 MRO team, in front of the 96 deployed antenna assemblies.

Picture6Figure 6. TPM subrack assemblies (four in total) deployed at the MRO CPF, including required peripheral equipment.

Initial test results of the 40-antenna subarray

Picture7Figure 7. Frequency response of 40 AAVS1 signal paths. System performance as expected, although a few channels seem to have lower power response.

Picture8Figure 8. Incoherent beam (summation of individual antenna power) of 16 antennas, showing increase in received power, as the Galaxy passes through the main beam. No calibration was applied.

Report provided by the LFAA consortium