Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.atnf.csiro.au/management/atuc/2005dec/docs/5.0_ATUC_TechDev_Report_dec05.doc Дата изменения: Fri Nov 25 07:20:39 2005 Дата индексирования: Sun Jun 27 16:31:41 2010 Кодировка: Поисковые слова: р п р п р п р п р п р п

ATNF ATUC Memorandum

To: ATUC
From: Warwick Wilson
Date: 24 November 2005
Subject Technology Development Report


ATCA 7mm systems

The contract with NASA to develop a 7mm (Ka-band) capability on the ATCA
has been finalised. It calls for the installation of the 7mm systems to be
completed by May 2007.

MNRF1997 - 3/12mm systems

The new Mopra MMIC-based 3/12mm receiver was installed in September. It
performed very well in a short observing session in October, with very good
Tsys from 77GHz to 115GHz. A spare ATCA 3/12mm frontend has been
completed. It is currently being used at Marsfield to do heat load tests
in preparation for the eventual installation of 7mm components in the ATCA
3/12mm receivers. When these tests are complete it will be sent to
Narrabri. The fabrication of a set of spare components for the ATCA and
Mopra 3/12mm receivers has been delayed but is expected to be completed by
March next year.

MNRF2001 - ATCA Broadband Upgrade (CABB)

Graham Carrad has taken over as project manager.

A major milestone was reached in October with the successful use of the
CABB prototype digital filter bank circuit board in a 2GHz-bandwidth
spectrometer (MOPS) at Mopra. This was a first demonstration of the
complex signal processing hardware and firmware that will be required for
CABB.

Based on experience gained from the development of the signal processing
system, and on the status of design work in other areas of CABB, a re-
evaluation of the overall project is being carried out. The result thus
far is a new implementation plan, which includes a more detailed model of
the installation and commissioning phase of the project and how this would
interact with other major activities at the time, such as the NASA 7mm
acceptance testing and spacecraft tracking. The new plan indicates a
significant overrun compared to the original June 2007 completion date for
CABB.

The new plan sees a "Stage 1" CABB becoming available for testing in July
07. Stage 1 will be a single frequency system with the antenna equipment
installed in such a way as to allow parallel operation of the existing
backend and the new CABB backend. The existing backend will be required to
provide Ka-band tracking for NASA, due to begin in July 07. The period
from July 07 until early 2008 will be used to thoroughly test the CABB
system. Testing will also be necessary to ensure that the CABB backend can
provide the required spacecraft tracking facility. Initial operational
use, with one frequency only, will also be possible during this time. At
the completion of this period an extended shutdown will be required to
remove the existing backend systems from the antennas and install the full
CABB antenna equipment in its final position.

A major issue recently has been uncertainty in the supply of the Xilinx
FPGAs which are a fundamental component of the overall CABB system design.
The immediate problem is a, at present indefinite, delay in the delivery of
FPGAs for the sampler prototype, originally due in October. There is
significant doubt whether Xilinx will ever achieve the performance required
for the current CABB design with this family of FPGAs. This has forced a
major re-think of the design of the sampler/data transmission systems.
Initial indications are that there are other practical options available.
There appear to be ways around the Xilinx problem in the DFB/Correlator
area.

Mopra 8GHz Spectrometer (MOPS)

After the successful installation and operation of one 2GHz quadrant of
MOPS, the major remaining tasks are the production fabrication of the DFB
PCBs and the design of firmware to provide the required performance in
terms of available bands and frequency resolutions. A detailed user
requirements document, taking into account limitations imposed by the now
well understood available processing power, is being prepared. Final
installation of the full 8GHz spectrometer is planned for May 06.

MNRF2001 - MMIC Development

Russell Gough has taken over as project manager.

InP MMICs

There will be a delay of around 6 months in the delivery of wafers from the
final InP fabrication run in the EU Faraday project collaboration, due to
an oversight by the manufacturer in obtaining export licenses. We are
attempting to have one wafer delivered in December. The most important
circuits required from this run are production quantities of the 7mm LNAs
and mixers.

Integrated Receivers

Devices from the initial fabrication run for the RF-CMOS integrated
receiver MMIC, containing RF test structures, were delivered in August.
All circuits, with the exception of the LNA, performed to design
specification. The problem with the LNA has since been identified. The
next run, containing more test structures and an initial full receiver
circuit will be submitted for fabrication in April 06.

6GHz Multibeam Receiver for Parkes

Installation of the receiver at Parkes has been delayed from November to
January 06 following a late change in the configuration of components
within the dewar. Jodrell Bank engineers decided to include isolators at
the inputs of their LNAs but did not inform ATNF. As a result, the LNA
mounting hardware, which had already been fabricated, needed to be
redesigned.

The receiver, without dewar internals, was taken to Parkes in November for
a test installation. This went very well and served its purpose in
revealing a number of minor issues that require attention before the final
installation.

Pulsar Digital Filterbank

The prototype 256MHz bandwidth pulsar digital filterbank was installed at
Parkes in June and has been operating very successfully since. The
production unit is largely complete, with the exception of the Xilinx
FPGAs, where the same problems with availability as mentioned above for
CABB are causing delays. In a worldwide search of suppliers, two chips
have been located and purchased for the signal processing system. The
chips are engineering prototypes, with reduced performance, allowing a
maximum bandwidth of 500MHz. Similar chips for the digitisers have not yet
been located. The project will be stalled until a source for these chips
can be found.



NTD

The test bed platform for the focal plane array technology is coming
together. The two ex-Fleurs dishes have been assembled, except for raising
the dishes onto the mounts. The dishes were assembled, surveyed, adjusted
and resurveyed. A new snake and mouse mesh surface has been installed. The
control systems have been built and tested. Fibre optic, coax and shielded
cat 5e cable have been bought and are ready for installation in the next
couple of weeks.

A modified THEA tile, which will be used as the first focal plane array
prototype, is being purchased from ASTRON. Douglas Hayman will travel to
Dwingeloo in the first week of December to witness the acceptance tests.

After several delays Mr Anatoliy Boryssenko has arrived from the University
of Massachusetts to assist the NTD antenna staff with the use of his
software to optimise the designs of phased array antenna feeds.

The development of the prototype digital beamformer and the 24 receivers
has progressed well. Both these systems have been packaged and tested
together. We have also been providing copies of these for the South
Africans. There has been communication with the South African KAT
development engineers and they are keen to collaborate to assist in our
further development.

The main activity for the next quarter is to commission and develop the
test bed and start experimental work assessing the performance of creating
beams using the THEA title as a phased focal plane array. The development
of other feed designs suitable for use in dense focal plane arrays is also
continuing. This work will be focussed on producing results for the NTD
critical design review (CDR) which is scheduled for the end of March 2006.

SKA Site Testing

The SKA International Radio Measurement team (ASTRON) finally made it on-
site at Mileura from September 23rd to October 19th. Their measurement
program went without a hitch and was successful. We understand that their
measurements agreed with ours and showed that Mileura is a world class
radio quiet site.

Measurements and data collection at Mileura continue as planned. The site
testing data collected thus far has been sampled and a first pass analysis
of the measured RFI has been completed. This was used in the preparation of
the site submission document. The SKA RFI Radio Trailer will be taken to
two "typical" remote array stations for a quick look survey as required by
the RFP.

Following detailed data processing, analysis and preparation of reports in
accordance with the requirements specified by the SKA Site Spectrum
Monitoring (SSSM) working group, a submission to the International SKA
Project Office is due on the 17th March 2006.