|
Since 1991, eight of the VLA antennas have been outfitted with
74 MHz receivers to permit testing the VLA's feasibility in imaging
at this low frequency. Although the tests were successful,
completion of the implementation was not possible due to lack of
funds.
Thanks to the Naval Research Lab, and the efforts of Namir
Kassim and Bill Erickson, funding for the completion of the VLA's
74 MHz receiver suite was secured in the spring of 1997. A new
design of Bill's was successfully tested over the summer, and
fabrication of 25 new receivers will be complete by the end of
September.
These new receivers are scheduled to be mounted in the remaining
twenty VLA antennas (there will be 5 spares for possible use with
the VLBA) during the fall and early winter of this year. Our plan
is to have the array completely outfitted by the end of the
D-configuration in January 1998, ready for use in the
A-configuration which starts at the end of that month.
The 74 MHz system will not be permanently mounted. Careful
measurements have demonstrated a significant loss in sensitivity of
the antennas at most of the other bands due to the presence of the
74 MHz dipoles. Until we can understand and correct the cause of
this loss of performance, the 74 MHz dipoles will be mounted only
during specific periods of time, during which all proposals for
that band will be scheduled. The first opportunity will be in
January and February of 1998, covering the end of D-configuration,
and the beginning of A-configuration. If there is sufficient
interest by the user community in VLA observations at this
frequency, the dipoles will be re-mounted in August or September of
1998 for use during the B and C configurations.
We are thus soliciting potential users of this band to
submit proposals by the normal deadline of 1 October, 1997, for the
A-configuration. Users must understand that this is a
trial band, and that calibration and imaging at this low frequency
where ionospheric effects are always very strong will be difficult.
The NRAO can not give any assurances that successful images of
weaker sources can be obtained.
Our previous tests have shown that the 'K-factor' (as defined in
the 'VLA Observational Status Summary') is about 1500, at
best. Since the system temperature is dominated by the galactic
background, and is typically in the range of 1000 to 3000 K, with a
maximum exceeding 7,000 K towards the galactic center, the
effective sensitivity of the array will be variable.
In practical terms, we expect that with a 12 hour integration
in a cold part of the sky, far from strong confusing sources, the
1-sigma noise may be as low as 20 mJy. We cannot give any
assurances that this impressive level will, in fact, be attained.
Our best images so far (using 8 antennas and 6 hours' integration)
have reached an r.m.s. noise level of about 100 mJy on 3C129.
Typical values on much stronger objects are a factor 2 to 3
greater.
The resolution of the array at this frequency in the
A-configuration is 25 arcseconds. The observational bandwidth is
limited to 1.6 MHz by the front-end filters, and it is strongly
recommended that observations be taken in spectral line mode to
permit calibration of instrumental delays and removal of any RFI.
Our previous tests, using 8 antennas, showed that
self-calibration at 74 MHz was straightforward, following the
methods well known at higher frequencies, providing an object
exceeding 50 Jy was in the isoplanatic patch of the target source.
In some cases, objects of much lower flux density may serve well --
a good image of 3C129, with a peak flux of only 11 Jy, was obtained
using these same techniques, although heroic measures were
required. Since the efficacy of self-calibration should increase
approximately linearly with the number of antennas, we expect that
standard self-calibration of objects with a peak flux of 15 Jy
should work well, and perhaps can be expected to work effectively
at levels much lower than this in colder, isolated regions of the
sky.
We have also tested a calibration scheme using phase transfer of
327 MHz data, taken simultaneously with the 74 MHz data. We have
shown that this method effectively removes the short-term
ionosphere, thus allowing subsequent self-calibration on a longer
time span. We are hopeful that this method should permit effective
use of this band for objects too weak for standard
self-calibration. Some testing, and software development will be
needed to make this technique effective.
Finally, note that RFI (both internal and external) is not a
serious problem at this frequency, as we are using a narrow
bandwidth of 1.6 MHz centered on the radio astronomy allocation of
73.8 MHz.
If you have any questions, please contact
Rick Perley (phone 505-835-7312).
|
National Radio
