The meeting was a teleconference, with participation from all NRAO sites. Several other NRAO staff participated as observers.
NB: The millimeter-wave array project is now a joint venture between the US and Europe, recently named "ALMA." However, the plans discussed at this meeting predate ALMA, and formal arrangements for European participation in such reviews have not yet been agreed. Accordingly, throughout this document reference is to MMA development rather than to ALMA. No major changes in holography planning are anticipated resulting from the increased scope of the ALMA project.
1. Requirements of MMA Holography
Darrel Emerson summarized the basic requirements:
The goals of measurements made on the prototype antennas:
The specific measurement requirements become:
Recommendation:
The specification should be kept at 10 microns,
but with a goal of reaching a precision of 5 microns.
To the requirement for a measurement set to be completed within one hour
should be added the goal of completing a measurement within 30 minutes.
Alternative measurement techniques:
Recommendation:
This technique may be convenient for the initial
antenna setting, where only 100 microns precision is needed.
For the prototype antennas, this is however defined to be the
responsibility of the antenna manufacturer. We should not rely on
photogrammetry for the final setting of the surface, although we do
need to keep up to date with developments in this field.
John Lugten outlined a commercially available laser metrology system using a movable retroreflector, which would be dragged over the antenna surface; see Laser CMM for Measuring the Primary Mirror Surface. Potentially this technique would be able to reach the required precision, but reservations expressed during the meeting included uncertainty in the magnitude of residual systematic effects, difficulties of maintaining continuity as the laser beams crossed obstacles such as the feed legs, and some degree of uncertainty as to how well the technique might perform in a relatively harsh exterior environment.
Recommendation:
We should watch developments in
the area of laser metrology. This may be a viable alternative to
holography. However, at the moment the panel does
not recommend abandoning the existing holographic measurement
plans in favor of such a scheme.
Other options: the possibility of using phase retrieval holography was discussed. This has the advantage of simplicity of hardware implementation; no special, dedicated holographic receiver would have to be constructed. However, experience at other telescopes indicates that, although phase retrievable holography is workable, the coherent holographic technique has significant advantages; it works well with much lower signal-to-noise ratios, and there is little room for ambiguity in interpretation of the results. It is perhaps significant that at least 2 major millimeter-wave observatories (the IRAM 30-meter, and the JCMT) had originally used phase retrieval holography, but have since opted to build and use coherent holographic systems.
Recommendation:
Stay with coherent holography as planned.
Richard Hills raised the question of whether we should be considering single-dish holography at all (see notes below on the different phases - single dish and interferometric - of holography implementation). The alternative is to wait until a complete, phase-stable interferometer system is available and to rely only on that. The biggest disadvantage is that the holographic measurements can then only be carried out after the full 2-element interferometric system has been sufficiently debugged and proven to be sufficiently phase stable.
Recommendation:
The advantages of having a single dish
holographic measurement system available before, and decoupled
from, the full interferometric system justify the construction
of the single-dish system as planned.
2. Overall plans for implementation of the MMA radio holography measurement scheme
Darrel Emerson outlined the plans. There are two distinct phases of holography:
The stability of the tower was of some concern. With a transmitter on a tower 300 m from the antenna, lateral movements of the tower should be kept to less than 1 mm. Although this stability may be achieved in calm conditions, some tower metrology system may be needed - for example laser measurements as demonstrated in GB for the GBT, or even a separate microwave interferometer using 2 horns mounted on the ground either side of the antenna being measured.
The choice of prime focus mounting of the holography receiver was questioned. There are pros and cons of both prime and secondary focus.
Reflections from the transmitter, its feed and enclosure may be a problem.
The near-field correction is not to be taken lightly, although there seems to be little alternative; if the tower is further from the transmitter, the maximum elevation angle becomes unacceptably small. The approximations involved in the correction become less if the antenna can be refocused on the tower; this requires a movement of at least 10 cm away from the far-field focus position of a prime-focus receiver.
The holography transmitter could also serve as an eventual test beacon for the full interferometric system, both on the US test site and at the final array location. For this purpose, it would be convenient later on to be able to incorporate a harmonic generator to give much greater wavelength coverage.
Recommendations:
The prime focus holography receiver and
near-field measurement using the tower-mounted transmitter is an
acceptable plan. Serious consideration should be given to arranging
for the holography system to have some
tunability, such as in 5 GHz steps over a 30% total bandwidth range.
Instabilities of the tower should be examined further,
and if necessary some form of metrology system to monitor
motions of the transmitter should be implemented. The
holography receiver mount must allow the receiver to be moved
into focus when observing the terrestrial transmitter. The effort
involved in installing and removing the prime focus holography
receiver should be minimized. The
transmitter design should take account of the need to minimize
spurious reflections. The importance of accurate characterisation
of the phase pattern of the prime focus holography feed needs some
emphasis.
3. Hardware implementation.
Antonio Perfetto presented detailed plans; see the material referenced at the top of this document, in particular Description of Hardware for Phase-Coherent Holography.
Discussion:
Some of the design specs appear to be overly
stringent. 120 dB for cross-talk is probably not necessary, and
there is no requirement for the high frequency stability. The
options of using a commercial vector voltmeter, vs. a simple
home-made correlator, were discussed. It was noted that in oder
to achieve an amplitude accuracy better than 1%, careful attention
must be paid to detector calibration. Care must be taken that the
transmitter does not interfere with the VLBA antenna at Pie Town,
and that it conforms strictly to all FCC regulations.
Recommendations:
The plans as presented are reasonable. For the
backend, the possibility of using an existing system should be
investigated - for example, that used for the SMT, or for the
GBT. Failing that, it may still be better to build a special
correlator (which might be as simple as an A/D and software)
rather than rely on the commercial vector voltmeter; this issue
should be studied further. More careful costing, both of
hardware and manpower, should be made; there was a feeling that
the estimated total cost ($160k) is probably somewhat too
high, while the included manpower estimates (3 man-months for each of one
engineer and one technician) may be too low.
[Correction:
At the meeting, $160k was presented as being
the materials-only cost. This was incorrect; 160k$ includes both
manpower and materials.] All in all the
hardware implementation plans are basically sound, although a
little more investigation of alternatives is required before
committing to the final design.
4. Real time and control software. Brian Glendenning presented the plans; a document Holography Software Development for the MMA had been made available in advance. The telescope control and data acquisition from the holography backend would be well integrated into the normal telescope control system.
Recommendations:
There was no disagreement on the plans as presented. Some suggestions included:
5. Data analysis software.
Brian Glendenning outlined the plans, which had been summarized in a document Holography support in AIPS++ distributed in advance. The data analysis, from gridding time-tagged holography data through to production of a telescope aperture error map, corrected for near-focus and other phase errors, and then to a detailed table of panel adjustments, would all be carried out within AIPS++.
Discussion:
It was suggested that the algorithm for deriving
the table of required panel adjustments might best be
implemented as an iterative calculation starting from assumed
surface errors, and comparing the raw measurements with
predictions from that surface model. The model would then be
adjusted for best fit to the raw data.
Recommendations:
Summary of recommendations:
The plans as presented are basically sound, although before committing to the final hardware design a little more study effort is required to consider some of the detailed design options. Specific points are covered in the individual recommendation sections.