South African Radio Astronomy Observatory

The SKA is a mega-science project, which will push the limits of engineering and scientific endeavour over the coming decades.

MeerKAT
The South African MeerKAT radio telescope, situated 90 km outside the small Northern Cape town of Carnarvon, is a precursor to the Square Kilometre Array (SKA) telescope and will be integrated into the mid-frequency component of SKA Phase 1.

MeerKAT’s make-up
The MeerKAT telescope is an array of 64 interlinked receptors (a receptor is the complete antenna structure, with the main reflector, sub-reflector and all receivers, digitisers and other electronics installed).

The configuration (placement) of the receptors is determined by the science objectives of the telescope.

48 of the receptors are concentrated in the core area which is approximately 1 km in diameter.

The longest distance between any two receptors (the so-called maximum baseline) is 8 km.

Each MeerKAT receptor consists of three main components:

The antenna positioner, which is a steerable dish on a pedestal;
A set of radio receivers;
A set of associated digitisers.
The antenna positioner is made up of the 13.5 m effective diameter main reflector, and a 3.8 m diameter sub-reflector. In this design, referred to as an ‘Offset Gregorian’ optical layout, there are no struts in the way to block or interrupt incoming electromagnetic signals. This ensures excellent optical performance, sensitivity and imaging quality, as well as good rejection of unwanted radio frequency interference from orbiting satellites and terrestrial radio transmitters. It also enables the installation of multiple receiver systems in the primary and secondary focal areas, and provides a number of other operational advantages.

The combined surface accuracy of the two reflectors is extremely high with a deviation from the ideal shape being no more than 0.6 mm RMS (root mean square). The main reflector surface is made up of 40 aluminium panels mounted on a steel support framework.

This framework is mounted on top of a yoke, which is in turn mounted on top of a pedestal. The combined height of the pedestal and yoke is just over 8 m. The height of the total structure is 19.5 m, and it weighs 42 tons.

The pedestal houses the antenna’s pointing control system.

Mounted at the top of the pedestal, beneath the yoke, are an azimuth drive and a geared azimuth bearing, which allow the main and sub-reflectors, together with the receiver indexer, to be rotated horizontally. The yoke houses the azimuth wrap, which guides all the cables when the antenna is rotated, and prevents them from becoming entangled or damaged. The structure allows an observation elevation range from 15 to 88 degrees, and an azimuth range from -185 degrees to +275 degrees, where north is at zero degrees.

The steerable antenna positioner can point the main reflector very accurately, to within 5 arcseconds (1.4 thousandths of a degree) under low-wind and night-time observing conditions, and to within 25 arcseconds (7 thousandths of a degree) during normal operational conditions.

How does MeerKAT work?
Electromagnetic waves from cosmic radio sources bounce off the main reflector, then off the sub-reflector, and are then focused in the feed horn, which is part of the receiver.

Each receptor can accommodate up to four receivers and digitisers mounted on the receiver indexer. The indexer is a rotating support structure that allows the appropriate receiver to be automatically moved into the antenna focus position, depending on the desired observation frequency.

The main function of the receiver is to capture the electromagnetic radiation and convert it to an voltage signal that is then amplified by cryogenic receivers that add very little noise to the signal. The first two receivers will be the L-Band and UHF Band Receivers.

Four digitisers will be mounted on the receiver indexer, close to the associated receivers. The function of the four digitisers is to convert the radio frequency (RF) voltage signal from the receiver into digital signals. This conversion is done by using an electronic component called an analogue to digital converter (ADC). The L-band digitiser samples at a rate of 1 712 million samples every second. (The amount of data that is generated by the digitiser for a receiver is equivalent to approximately 73 000 DVDs every day or almost 1 DVD per sec.)

Once the signal is converted to digital data, the digitiser sends this data via buried fibre optic cables to the correlator, which is situated inside the Karoo Array Processor Building (KAPB) at the Losberg site complex.

A total of 170 km of buried fibre cables connect the receptors to the KAPB, with the maximum length between the KAPB and a single antenna being 12 km.

The fibre cables run inside conduits buried 1 m below the ground for thermal stability.

At the KAPB, the signals undergo various stages of digital processing, such as correlation – which combines all the signals from all the receptors to form an image of the area of the sky to which the antennas are pointing – and beam-forming, which coherently adds the signals from all the receptors to form a number of narrow, high sensitivity beams used for pulsar science. The science data products are also archived at the KAPB with a portion of the science archive data moved off site via fibre connection and stored in Cape Town (with possibilities of reprocessing the data).

Time and frequency reference signals are distributed, via buried optical fibres, to every digitiser on every receptor, so that they are all synchronised to the same clock. This is important to properly align the signals from all receptors.

The control and monitoring system is responsible for monitoring the health of the telescope and for controlling it to do what the operators want it to do. A large number of internal sensors (more than 150 000) monitor everything from electronic component temperatures to weather conditions and power consumption.

Why MeerKAT?
The telescope was originally known as the Karoo Array Telescope (KAT) that would consist of 20 receptors. When the South African government increased the budget to allow the building of 64 receptors, the team re-named it “MeerKAT” – ie “more of KAT”. The MeerKAT (scientific name Suricata suricatta) is also a much beloved small mammal that lives in the Karoo region.
---

KAT-7
The seven-dish MeerKAT precursor array, KAT-7, is the world’s first radio telescope array consisting of composite antenna structures.

KAT-7 was primarily built as a precursor to the 64-dish MeerKAT radio telescope array and to demonstrate South Africa’s ability to host the SKA. However, it has proved to be a pioneering scientific instrument in its own right.

KAT-7 is considered a compact radio telescope, since its antennas all lie within an area only 200m across, as opposed to the much larger areas that will be occupied by MeerKAT and the SKA. The KAT-7 configuration is perfect for observing nearby galaxies, which emit radio waves on a large scale.

How it works
Each of its antennas consists of a movable dish mounted on top of a supporting pedestal, which houses cables and some components. The dishes measure 12 metres in diameter and, in a world first, are made entirely out of fibre glass.

The curved surface of the dish reflects incoming radio waves from space to the receiver, held above the centre of the dish by metal rods. The dish focuses the radio signals into the feedhorn of the receiver in the centre of the dish.

These radio receivers, which pick up signals in the frequency range of 1200–1950 MHz, are cooled to about 70 degree Kelvin (-203°C) in order to reduce the “noise” inherent in all radio receivers. At this temperature, the receiver is 2.5 times more sensitive than at ambient temperatures, allowing images to be captured roughly six times faster. This in turn allows for the detection of much fainter celestial objects.

Once captured by the receiver, the radio signals, which are still in analogue form, travel via fibre optical cables from the antennas to the Karoo Array Processor Building (KAPB), where they are digitised and correlated.

Lessons for MeerKAT
As a pathfinder for MeerKAT, engineers used KAT-7 to test, develop or improve upon many of the technologies being implemented in the larger array.

While KAT-7’s dishes are constructed in a more traditional way, with the receiver directly above the centre of the dish, MeerKAT’s dishes are in a so-called offset Gregorian configuration. This simply means that the dish itself has a slightly different curvature, so as to reflect radio waves to the receiver held out in front of it, nearer to the bottom of the dish, by a single metal arm. Also on that arm is a rotating platform that can house up to four receivers (covering different radio frequencies) and their corresponding digitisers. Whereas digitisation of KAT- 7 data occurs at the KAPB just before correlation, MeerKAT digitisation occurs on the antenna itself, before the data are sent along fibre optics to the KAPB for correlation.

Up-close, a KAT-7 antenna hums a near constant tune due to the ion pump that cools the receiver in a manner similar to a household refrigerator. In contrast, MeerKAT’s receivers are cryogenically cooled using a much quieter helium compressor.

KAT-7 science
Astronomers have used KAT-7 to monitor Circinus X-1, which is thought to be a binary start system. KAT-7 has also peered into a small region known as the Hubble Ultra Deep Field, which is where the deepest optical observation was made with the Hubble Space Telescope. In addition, the array has also observed radio signals associated with hydrogen emission from a nearby galaxy (NGC 3109) for the first time, as well as the blazar called PKS 1510-089.

---

HERA
HERA (Hydrogen Epoch of Reionization Array radio telescope) will attempt to answer one of the biggest scientific questions today: how did the universe come to be?

The HERA (Hydrogen Epoch of Reionisation Array) radio telescope will be instrumental in detecting the distinctive signature that would allow astronomers to understand the formation and evolution of the very first luminous sources: the first stars and galaxies in the Universe.
To solve the mystery of how we came to be where we are, astronomers must look back in time to a transitional period of cosmic history that has been dubbed the “epoch of reionisation”. Nobody has yet detected a signal from this period, but the Precision Array for Probing the Epoch of Reionisation (PAPER) pathfinder instrument has come the closest so far. HERA is expected to measure such signal throughout the epoch of reionization.

What is HERA
HERA is an American, South African and British collaboration to build a telescope capable of making a solid detection of the Epoch Of Reionisation (EOR) red-shifted hydrogen power spectrum signature, as well as conducting initial EOR science and launching this new scientific field of the observational cosmic dawn. The discovery and the early advancement of this science is a likely candidate for a Nobel Prize.

How HERA works
Unlike KAT-7, MeerKAT and eventually the SKA, which are general-purpose instruments intended to perform many different scientific observations, HERA has just one goal: to characterise the epoch of reionisation. It will give us a 3D map of the universe during this era. HERA is designed to detect radio waves in the low-frequency range of 100–200 MHz, which allows it to detect fluctuations in the emissions from neutral hydrogen gas that was found throughout the universe before stars, galaxies and black holes formed. Being a low-frequency instrument, the field-of-view for an antenna of a particular size is much larger than it would be if high frequencies were being detected. In other words, one can see more of the sky in one go at low frequencies.

This, along with the fact that the sought signal is everywhere in the sky (much like the Cosmic Microwave Background signal), means that individual antennas do not have to move around or be pointed at specific locations to scan for signals. No expensive moving parts are thus required. Whereas MeerKAT and SKA dishes rotate both vertically and horizontally to survey the sky, HERA antennas are immobile, pointing straight up at all times.

“HERA is expected to be the most sensitive SKA pathfinder to study the EOR, the period when the first galaxies were formed and started to shine – a cutting edge research field in modern cosmology and one of the main science cases for the SKA. We expect that HERA will take the field one leap forward and will be able to inform the SKA in terms of searching for the HI signal at very high redshift,” explains Dr Gianni Bernardi, SKA SA Senior Astronomer.

PAPER and HERA
HERA is located in the South African Karoo Astronomy Reserve, with a nominal array centre of 30°43’17″S and 21°25’42″E – the location of the existing Precision Array for Probing the Epoch of Reionisation (PAPER).

PAPER listens in on both the Northern and Southern hemisphere skies. The primary 128-antenna instrument is at the SKA site in the Karoo, while its smaller 32-antenna cousin is in West Virginia in the United States.

While PAPER is a pathfinder experiment (also known as HERA phase I) consisting of small antennas only a few metres across, each of HERA’s 331 antennas will measure 14 metres in diameter and boast a collecting area roughly 30 times larger. Bigger antennas mean that HERA will have a smaller field-of-view than PAPER, but it will be much more sensitive to the faint signals from the epoch of reionisation. Each antenna measures dual polarisation and they are referred to as dipole antennas.

HERA works on the same interferometry principles as KAT-7, MeerKAT and the SKA: essentially, data from all its antennas are combined. The instrument also shares much of the supporting infrastructure being constructed for MeerKAT and the SKA, including the underlying technology of the digital back-end, which processes and packages the incoming analogue radio signals so that astronomers can glean information and create visual images using digital data.

The current PAPER container will be moved to a location just west of HERA and the correlator will be moved to the Karoo Array Processor Building (KAPB). PAPER will be decommissioned over the HERA construction period.

HERA collaboration under way
The US National Science Foundation is the primary funder of the current HERA prototype phase, with significant funds from the University of Cambridge Cavendish Laboratory, the University of California at Berkeley and the National Radio Astronomy Observatory, with site and logistical support from SKA South Africa.

Construction will be sourced and constructed from within South Africa – predominantly from the Carnarvon area. South African scientists from SKA SA and numerous South African universities (the University of the Witwatersrand, University of KwaZulu-Natal, University of Cape Town and the University of the Western Cape) are significantly involved with the science, and researchers and engineers from SKA SA, the Stellenbosch University and the Durban University of Technology are expected to be heavily involved in the design and production of the array. Rhodes University, through the Research Associate Dr Gianni Bernardi, is also involved in research using PAPER and HERA.

PAPER (HERA phase I)
PAPER has collected data for five years, with the aim of establishing how strongly the signal from the Epoch Of Reionisation (EOR) can be detected using the pioneering approach planned for HERA.
PAPER has produced the best upper limits on the EOR signal to date.
Also thanks to PAPER, scientists now have a roadmap for the next phases of HERA, directing them along the best routes for configuring the array, processing and storing the data, and preventing interference from other radio wave sources that would distort the signal.

---

C-BASS
C-BASS (C-Band All Sky Survey project) is a project to map the sky in microwave (short-wavelength radio) radiation.