The SKA Molonglo Prototype (SKAMP)
Prototyping SKA Technologies at the Molonglo Radio Telescope
The Square Kilometre Array (SKA) is an international project to build a radio telescope which will be 100 times more sensitive than any existing radio telescope with about one million square metres of collecting area. In order to design and build this new generation radio telescope major new technology is needed. SKAMP is a joint CSIRO/University of Sydney project, largely funded through the Australian Government's Major National Research Facilities (MNRF) program, to develop and test new technology for the SKA.
The telescope at Molonglo will be equipped with new wide-band feeds, low-noise amplifiers, digital filterbanks and FX correlator, and demonstrate 300-1420 MHz continuous frequency coverage and multibeam mode operation. This will allow us to develop new capabilites for low-frequency radio astronomy in Australia, enabling exploration of the distant universe.
Collaboration at CSIRO is with engineering and science teams at the Australia Telescope National Facility (ATNF) and the ICT division of Wireless Technology. This collaboration is integral to the success of the project; the final design and implementation will be jointly achieved.
A Powerpoint overview of the Molonglo Observatory and the SKAMP projects available.
The following image illustrates the three key stages to the SKAMP project.
Specifications of SKAMP
|Field of View||2.2 deg FWHM EW x 2.2 cosec |dec| deg FWHM NS approx at 843MHz, i.e. the FoV stretches NS as declination moves from the pole towards the equator.|
|System Temperature||80K (Tsky=10K, LNA T=20K, Feed Loss= 10K, Ground spill = 10K, Mesh leak = 15K, Tmatch = 15K)|
|Geometric Area||11.6m * 1556m = 18,000sq m|
|Illumination Efficiency||.5 (Mesh efficiency ~ .95, MD projection efficiency, unity at SCP then efficiency will go up and down ~ .5, MD alignment efficiency ~ .95)|
|Effective Area||.45 * 18,000 = 8100 sq m (this will range from a max of 8100 at the south pole to around half this at dec -30)|
|Elevation Limit||35deg (North South), 30deg (East West)|
|Observing Frequency||SKAMP II - 843MHz. SKAMP III - 700-1400 MHz in 100 MHz bands.|
|Noise in 12 hr observation as:|| (a) A Continuum Correlator: 0.2 mJy rms(SKAMP 2 estimated, based on BW of 10MHz effective due to RFI); 0.1 mJy rms (SKAMP 3 estimated). Note that for sources >0.8mJy the source density is 130/deg^2, corresponding to 50 beam areas/source, close to the confusion limit.
|Angular resolution||43 x 43 cosec |dec| arc seconds|
|Bandwidth||SKAMP II - 30MHz (nominal, 50% response; limited by bandwidth of ring antennas). SKAMP III - 93.75MHz|
|Data Rate||.75Gb/s, 32 sec dump time, corresponding to 337 GB/hr.|
|# of channels||6480 channels (14.4kHz channel spacing; 5.1 km/s at 843 MHz) obtained by averaging over 4 channels of 3.61 kHz.|
|# of Receivers||SKAMP II - 352 (88*4). SKAMP III - 176 for each polarization.|
|# of Polarizations||SKAMP II - 1 (elliptical right circular polarization). SKAMP III - 2 (linear)|
|Operational Modes||Limited continuum observations (1,2,4,8,10 MHz, spectral line at 15MHz); 12 hour observation.|
SKAMP III - The RPT, with meshing complete, is now up on its stand and secured in position. One Antenna module has been installed - as pictured.
Frontend & Antennas:
- December 2010 - Final prototypes of receiver FPGA board and down converter received.
- December 2009 - Spectrum obtained during coarse spectrum testing.
- October 2009 - Version 2 of the Receiver FPGA board and Down Converter complete and undergoing testing.
- Information on SKAMP Antennas and frontend prior to 2007 can be found here
SKAMP Continuum Correlator and Digital Systems:
- February 2011 - All FPGAs on correlator board have been exercised.
- December 2010 - Functional testing of correlation cells along with the cell control system resident in the CMAC now complete.
- December 2010 - Digital System hardware manufactured by CSIRO. Final testing underway in preparation for delivery to SKAMP on 4th January 2011
- September 2010 - SKAMP Digital System Handover Event.
- February 2010 - Preliminary design review meeting for SKAMP correlator Firmware held to ensure development plan on track and specifications are met.
- November 2009 - Darshan Thakkar, Duncan Campbell-Wilson and Ludi de Souza are help with the end-to-end testing and integration of the MWA digital system at Curtin.
- March 2009 - Boards are ready for production. Correlator firmware is currently under development.
- Project status prior to 2007 can be found following the link below
Data Acquisition and Software Pipeline:
- February 2011 - SKUMP: a set of visibilities from all antennas at a particular frequency for one unit of time. This term was developed to describe the SKAMP data dump from the correlator board (approx 70,000 complex numbers)
- January 2011 - Greg Madsen and Jay Banyer join the SKAMP team to work on Data Acquisition & the Software Pipeline.
- November 2010 - Software team meeting discusses progress on the TCC and Software Pipeline. SKAMP's scientific goals drive the software development and will be the focus of several people over the coming months.
- Data pipeline software now available for testing.
To view all SKAMP Newsletters from August 2010, click here.
Latest SKAMP news
Its a telescope - but not as we know it Deborah Smith, SMH, 2 June 2011
Some say it would be like winning a bid for the Olympics. And if Australia is chosen early next year as the site for the world's largest telescope, there will be cause for similar celebration. Read more ...
Hip hip hooray for the (Aussie?) Square Kilometre Array: by Bryan Gaensler.
April 2011. Bryan Gaensler, Australian Laureate Fellow; and Director of CAASTRO at University of Sydney has written an article on the Square Kilometre Array for The Conversation.
The Conversation is a new Melbourne-based website that has been created by the Group of Eight Universities, of which the University of Sydney is one, to encourage academics and students to write articles on their area of research and expertise.
SKAMP NEWSLETTER - February 2011 issue
The latest issue of the SKAMP NEWSLETTER is now available!
January 2011: The SKAMP team held a one day workshop, designed to introduce new members to the project, review the work that has been done and set project deliverables for 2011. It also provided the opportunity to discuss the contribution that SKAMP will be making to a number of science projects linked to SKA (Square Kilometre Array) key goals. Following presentations on Science with Radio Astronomy and The History of the Molonglo Observatory, the discussions focused more specifically on SKAMP progress, our links with CAASTRO (Centre for All-Sky Astrophysics) and the transient science that is currently driving the project.
Also discussed were further science projects related to SKA goals, in which SKAMP will play a role. These included: Blind survey of HI absorption in high redshift galaxies (z~0.7); Search for OH megamasers in disks around super-massive black holes (z~1); Cosmic magnetism studies – “diffuse Galactic polarisation and Rotation Measure”; and Radio Recombination Lines (H199α is 841 MHz).
The meeting concluded with in-depth discussions on the software needs of SKAMP and transient research. Immediate priorities were set for the software development team with a follow-up meeting held on-site at the Molonglo Observatory to finalise the development strategy.
The SKAMP digital system hardware is now complete and the front end receivers are nearing production, so efforts are really beginning to focus on the design and development of the front end and back end software. Contributions from all workshop attendees ensured the meeting was a success. In attendance: Keith Bannister, Jay Banyer, Duncan Campbell-Wilson, Sean Farrell, Bryan Gaensler, Deb Gooley, Anne Green, Paul Hancock, Dick Hunstead, Mike Kesteven (CSIRO), Kitty Lo, Greg Madsen, Tara Murphy and Darshan Thakkar.
In the Media
November 2010: Revamped SKAMP boosts Australia’s chances of hosting the SKA.
www.cio.com.au (Diana Nguyen, Computerworld).
The revitalisation of the Square Kilometre Array Molonglo Prototype (SKAMP), a joint effort between the University of Sydney and Australia’s Commonwealth Scientific and Industrial Research Organisation (CSIRO), is expected to help further Australia’s bid to host the Square Kilometre Array (SKA) telescope over rival South Africa . . . More.
SKAMP Digital Systems Handover
September 2010: Making the MOST of SKAMP.
The Sydney Institute for Astronomy (SIfA) based within the School of Physics, and CSIRO have collaborated in a project that will transform an existing university radio telescope into a world-class instrument. Using CSIRO technology it will be built into a new international telescope, and shape the design of its own ASKAP (Australian SKA Precursor) telescope. All three telescopes will demonstrate technology for the SKA, the international Square Kilometre Array radio telescope. More.
SKAMP designated official SKA Pathfinder status
August 2010: In early August the SKA Science and Engineering Committee (SSEC) designated SKAMP as an approved SKA Pathfinder. The SSEC is the sole arbiter for designating national, regional and institutional contributions to the SKA as Precursor, Pathfinder, or Design Study.
Following a proposal submission by Anne Green, SKAMP Project Leader, the SSEC determined that SKAMP’s contribution satisfied the Pathfinder criteria within the areas of technology, science and operations. SKA Pathfinder: SKA-related technology, science & operations activity. For more information on SKA designations, visit www.skatelescope.org.
AnzSKA Newsletter contribution
Apr 2010:The prototype SKAMP receiver has been developed to a point where it achieves its primary goal: it is frequency agile, and operating all data, clocks, local oscillators and control signals on optic fibre. The prototype receiver is also producing a frequency spectrum consistent with known local carrier signals. Managing large receiver numbers (352) has been designed into the system for in-field diagnostics, and the prototype receiver is now being prepared and configured for field operations.
The polyphase filterbank system connects and locks to the receiver data streams, and the coarse frequency spectrum that is being produced is consistent with carriers in the local environment. Development work to achieve the fine spectral resolution required by the second stage of SKAMP is also continuing.
Correlator development is progressing on two paths: a 16-channel correlator with firmware has been designed by MIT Haystack for 16-channel end-to-end tests, and a second path is being developed by the University of Sydney to meet the SKAMP correlator performance requirements within the hardware design constraints. The latter is showing good progress in both processing speed and resource utilisation.
The digital backend in the shielded room built for SKAMP at the Molonglo Radio Observatory. Photo courtesy of Adrian Blake, USYD
WARS 2010 - SKAMP
Feb 2010: The SKAMP system details were presented at the poster session of WARS 2010, Canberra, 11th-12th February 2010. Please click on the following link to access the poster, SKAMP: Re-engineering the Molonglo Radio Telescope (jpg 2.35MB), D. Campbell-Wilson "et. al."
Feb 2010: The development of the SKAMP correlator firmware is now well underway. The local development team, Darshan Thakkar, Ludi de Souza and Duncan Campell-Wilson held a preliminary design review meeting this month. The correlator cell and control system have been functionally tested and the team is now focusing on the CMAC architecture.
SKAMP Receiver-PFB Link Testing
Dec 2009: The SKAMP team at the Molonglo Observatory have run signals through the receiver and polyphase filterbanks, achieving the following spectrums at the coarse PFB output stage.
Spectrum of F2, 18th Dec 09. Courtesy of Ludi de Souza, CASS
Spectrum - No F2, 18th Dec 09. Courtesy of Ludi de Souza, CASS
Spectrum of Rx signals through PFB, 18th Dec 09. Courtesy of Ludi de Souza, CASS
Oct 2009: Duncan Campbell-Wilson and Darshan Thakkar have been making good progress on the development of the receiver and downconverter boards for the SKAMP front end. Version 2 of the receiver board (Rx V2) has been received and FPGA placed. The installed local oscillator (LO) distribution boards are now being tested in conjunction with the Rx V2. The primary objective of this testing will be to produce reliable clock and LO signals with minimum distortion, noise and mixing products.
Bit Error Rate Tests on the Receiver-PFB interface have been carried out successfully, and complete path testing will soon follow. Advanced diagnostics are being added to the Receiver Firmware and Software for sophisticated remote debugging.
Version 2 of the Down Converter prototype has also been received. It has been powered up and is currently being tested.
FPGA Receiver Board. Photos courtesy Molonglo Observatory
FPGA Rx board and Down Converter boards within the Rx Box.
The brass boxes over the Down Converter minimize mutual coupling between input channels.
Science with SKAMP: Widefield Spectroscopy of the Southern Radio Sky - September 2009
A highly successful two day meeting, "Science with SKAMP: Widefield Spectroscopy of the Southern Radio Sky", was held at Molonglo Observatory on 16-17 September, 2009. Approximately 40 astronomers participated in this meeting, including international speakers from the USA, Canada and India.
Topics covered during the meeting included:
- The status of SKAMP instrumentation;
- Surveys for redshifted 21-cm and OH absorption systems;
- Searches for variables and transients;
- Studies of radio recombination lines in the interstellar medium;
- Mapping of non-thermal emission and magnetic fields in the Milky Way
and other galaxies;
- Synergies with other upcoming wide-field surveys with the MWA, SkyMapper
Various upcoming challenges for SKAMP surveys were discussed: these include the development of a real-time pipeline needed to handle the large data rate, the hostile radio interference environment resulting from SKAMP's proximity to Canberra and source confusion in deep integrations.
On the other hand, it is clear that SKAMP has some unique capabilities amongst SKA pathfinders: it has a large collecting area already in place, offers continuous and highly redundant u-v coverage, is a university instrument that can be dedicated to large projects, and already has all-sky reference images in hand from previous sky surveys. The presentations and ensuing discussion at this stimulating workshop made clear that there is broad community interest in the exciting scientific opportunities that SKAMP will soon be offering.
The workshop programme can be found here.
Copy courtesy of Bryan Gaensler, email@example.com
Click here to view larger image.
Science with SKAMP workshop delegates at the Molonglo Radio Observatory
SKAMP & MWA teams at ATNF, CSIRO - Aug/Sept 2009
Darshan Thakkar, from the Molonglo Observatory and Russ McWhirter, from the MIT Haystack Observatory travelled to Sydney early August to begin a period of intensive firmware development for the SKAMP and MWA projects at ATNF, CSIRO. Working closely with Ludi de Souza and John Bunton from ATNF, Darshan and Russ have been finalising and testing the PFB firmware as well as working with Bart Kincaid (MIT Haystack) to complete the MWA 32T correlator firmware. A full set of boards for the SKAMP system, including receiver was set up at ATNF for testing.
During the final week, the firmware teams from ATNF, MIT, The University of Sydney and RRI travelled to Curtin Univ to carry out the hardware integration activities. Testing involved ensuring data integrity and functional testing from the antennas, beamformers, MWA receiver boards (ADFB and AGFO), fine PFB board, correlator board through to the Real Time Computer that process the signals into an image cube. Hardware was transported from MIT for integration, and some damage was caused to the card cages and correlator. However, these were quickly rectified and end-to-end testing progressed well throughout the week.
Significant progress has also been made on the SKAMP receiver over the last few weeks. Duncan Campbell-Wilson successfully sent a sky signal through the receiver to the PFB to produce a spectrum.
Integration team at Curtin. Front: Srivani KS (RRI), Darshan Thakkar (USYD), Ludi de Souza (ATNF), Prabu T (RRI); Back: Wayne Arcus (Curtin), Russ McWhirter (MIT), David Emrich (Curtin)
Molonglo Renovations - May 2009
During May/June 2009, the main building at the Molonglo observatory underwent some major renovation in preparation for SKAMP II, in particular to improve the RFI shielding required for the project. The control room has been partitioned to create a new RFI shielded, sound-proof room for the SKAMP II digital system. This room is almost complete and only requires sound-proofing.
SKAMP II needs to be insulated from noise up to 50DB. After several discussions, multi-layer approach to RFI shielding was considered the most effective and cost-efficient way forward:
- Double mesh on the telescope
- Mesh over the outside of the building which extends underground
- Metal cladding over the southern part of the building
- Shielded SKAMP II room within the control room.
The new partitioned room for SKAMP II - not quite complete.
Asbestos being removed in preparation for re-cladding
Farewell to the TCC Console
In January 2009, Molonglo bid farewell to the famous telescope control computer console which has been a much admired fixture since the 1960s. The Correlator and Digital Systems for SKAMP will be installed within ATCA cages in the control room at Molonglo, which will need their own sound-proof room. This leaves little room for the TCC console.
The original TCC console at Molonglo
TCC now installed within these racks
SKAMP Digital Systems
The Digital System for SKAMP has just been assessed via the Production Readiness Review and will be going into production mid-2009. CSIRO, through John Bunton, and MIT Haystack, through Dr Roger Cappallo, have developed the architectural design of the Digital System.
The Digital System for SKAMP is made up of the following boards which will all be housed within ATCA cages at Molonglo:
- Correlator Board x 24: Firmware is currently under development at MIT, Haystack
- Polyphase Filterbank Board (PFB) x 24: Firmware developed by Ludi de Souza and is now complete.
- Slot-1 Controller x 4
- Correlator RTM x 24
- Polyphase Filterbank RTM x 24
Correlator board prototype (August 2008)
PFB board prototype (August 2008)
- Correlator and PFB photos courtesy of CSIRO.
NEW RELEASE - Nov 2008 - SKAMP Data Pipeline Software. Version 1.11a of this new software created by Michael Birchall is now available for testing.
Prior to 2008
SKAMP - Archive of Project Updates. Below - archives in brief:
- SKAMP I - First 12-hour image!
- SKAMP team contracted Domain-42 Pty. Ltd. to assist with the design of the spectral line correlator (SKAMP II).
- Xilinx donates over $30,000 worth of FPGAs.
- Power replaced and rewired at Molonglo Observatory in readiness for new correlator.
- Good progress on design of new wideband receiver and digitiser required for SKAMP II.
- Argus Technology support and fund PhD student to develop dual polarised line feed for SKAMP III.
News of other SKA-related projects in Australia and New Zealand
See the latest Australian SKA Newsletter.
Recent meetings and events
News and details of international meetings can be found on the International SKA webpage.
- Anne Green (USyd Project Director)
- Debra Gooley (SKAMP Project Coordinator)
- Bryan Gaensler (Federation Fellow)
- Tara Murphy (Science Advisor)
- Michael Kesteven (Project Scientist/Software)
- John Bunton (Correlator Systems Manager)
- Elaine Sadler (Science Advisor)
- Tim Adams (Consultant)
- Ludi deSouza (Digital Systems Design)
- Greg Madsen (Science Advisor/Liaison)
- Jay Banyer (Software Engineer)
- Martin Leung (Antennas)
- Alan Ng (CSIRO)
- John Tuthill (CSIRO)
- Ron Koenig (CSIRO)
- Michael Birchall (AAO)
- John Russell (UTas)
- Chris Weimann (UTas)
- Duncan Campbell-Wilson (Observatory Manager/Front-end Electronics)
- John Wills (Senior Technical Officer)
- Darshan Thakkar (Digital Engineer)
- Lindsay Harkness