Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.atnf.csiro.au/people/Michael.Kesteven/SURVEY/nbr_report.ps Дата изменения: Thu Oct 5 16:37:53 2000 Дата индексирования: Tue Oct 2 18:50:07 2012 Кодировка: Поисковые слова: arp 220

ATNF antenna optics
M.Kesteven & B. Parsons
April 26, 2000
As part of the MNRF upgrade of the Narrabri antennas, we have started a program of careful mea-
surements to ensure that optics are correctly aligned.
Our starting point is the main re ector surface, adjusted following the holography measurements made
in October, 1999.
We are now working through each antenna in turn, to dene in a permanent fashion, the optical axis
of the main re ector; we can then position the feeds and the subre ector on the optical axis, at the
correct distance from the re ector vertex.
We nd that in all the antennas surveyed the receivers have been installed with precision - they are
all within an error circle of radius 10mm. Only at the highest frequencies (80 GHz and above) will
greater precision be required, as far as eфciency is concerned.
The osets do have some consequences for other frequencies:
- Pointing Dierent receivers will have dierent squint terms in the pointing model (the pointing
error is 3.6 arcsec/mm for feed osets).
- Subre ector SettingThe holography data contain an error signature of the combined (sub-re ector
+ feed osets). We need to ensure that both the holography receiver and the mm receiver have the
same oset (preferably zero).
1 Background
An antenna operates at peak eфciency when all the optical elements (main re ector, sub-re ector and
feed) are at their designed locations. For the AT antennas, the subre ector vertex should be 6.430mm
above the main re ector vertex; the feed should be 1.509mm above the vertex. Departures from the
nominal alignment will reduce the antenna's performance. Figures 1 and 2 show the subre ector focus
curves; the corresponding curves for feed osets are similar in shape, but scaled by the magnication
factor (M=26 in the axial direction, and M=7 in the lateral direction). Since the losses result from a
phase distribution over the aperture plane, we can counter a feed oset with a subre ector adjustment,
although we can never fully recover all the gain.
However, all the feeds in an antenna should have the similar osets since the AT has no mechanised
lateral adjustment of the subre ector.
Holography is a powerful tool to measure the alignment of the optics; however it is diфcult to distin-
guish between subre ector and feed osets since their error signatures are similar. The saving grace
is the scale: a feed oset
x is eectively equivalent to a subre ector oset
x=M , where M, the
magnication, is about 7.0 for the AT. If we can start with feed and subre ector set with comparable
accuracies, then the holography will show primarily the subre ector error.
1

The work described in this note emphasises this approach: we provide the basis to ensure that the
feeds are accurately positioned. We also ensure that the subre ectors are close to their designed
location, and assume that holography/pattern-cuts will be used to rene the subre ector setting.
Figure 1: The normalised axial focussing curve for the AT antennas
2 The Reference Jig
We have built a reference beam for the AT antennas which, when installed on precisely dened locating
pins in the vertex room roof, locate the optical axis of the antenna.
The dowel pins were installed in the Mopra antenna in January, and in CA03, CA04 and CA02 in
April.
3 The Permanent Targets
We have also provided a set of permanent targets on the main re ecting surface which can be used to
determine the location and orientation of the subre ector relative to the main re ector.
The targets are 1/16 holes drilled in the panel surface, at the eight cardinal points, at the inner edge
of ring 1, and at the outer edges of rings 5 and 6.
2

Figure 2: The normalised lateral (in the focal plane) focussing curve for the AT antennas
4 Measurements
Errors We estimate the accuracy associated with these results to be 2mm.
TERMINOLOGY We refer to positions on the re ector in terms of "compass" bearings - if the
antenna were pointing towards the horizon, and you were looking towards the surface, North is the
highest point, East is to your right, South is the lowest point, and West is to your left.
4.1 CA02
The location of the various receiver packages is given in Table 1, and shown in gure 3
Receiver E-W oset (mm) N-S oset (mm) notes
C/X -1.0 +2.3
L/S +1.0 +11.0
Table 1: Antenna CA02 - receiver locations, relative to the main re ector optical axis
Subre ector
- The subre ector's axis is parallel to the main re ector axis to within 0.5 arcmin (NS) and 2.6 arcmin
(EW).
- The subre ector's axis is oset from the optical axis 1mm towards the North, and 1mm towards the
3

5 mm
N
L/S feed horn
Optical axis
C/X feed horn
Figure 3: Antenna CA02 - receiver locations, relative to the main re ector optical axis
East.
The Targets
radius(mm) height above vertex (mm)
R1 (inner) 1889 133.1
R5 (outer) 9338 3147.8
R6 (outer) 10937 4279.1
The lower surface of the optical-axis locating jig is 1522 mm above the vertex.
4.2 CA03
The location of the various receiver packages is given in Table 2, and shown in gure 4
Receiver E-W oset (mm) N-S oset (mm) notes
22 GHz -8.5 -7.0
C/X +7.5 -1.25 (before index adjustment)
+0.5 -1.25 (after index adjustment)
L/S +5.0 -10.0
Table 2: Antenna CA03 - receiver locations, relative to the main re ector optical axis
22 GHz receiver
The face of the 22 GHz feed horn (below the clamping ring) is 1450 mm above the R1 (inner) targets.
subre ector
- The subre ector's axis is parallel to the main re ector axis to within 2.5 arcmin (NS) and 1.24 mm
(EW).
- The subre ector's axis is oset from the optical axis 1mm towards the East
The targets
4

C/X feed horn
K­band feed horn
L/S feed horn
Optical axis
5 mm
N
Figure 4: Antenna CA03 - receiver locations, relative to the main re ector optical axis
radius(mm) height above vertex (mm)
R1 (inner) 1841 127.1
R5 (outer) 9267 3101.0
R6 (outer) 10936 4278.4
The lower surface of the optical-axis locating jig is 1515 mm above the vertex.
4.3 CA04
The location of the various receiver packages is given in Table 3, and shown in gure 5
Receiver E-W oset (mm) N-S oset (mm) notes
22 GHz -3.0 +5.0
C/X -3.3 -6.3
L/S +4.0 -0.7
Table 3: Antenna CA04 - receiver locations, relative to the main re ector optical axis
5 mm
N
Optical axis
L/S feed horn
C/X feed horn
K­band feed horn
Figure 5: Antenna CA04 - receiver locations, relative to the main re ector optical axis
5

22 GHz receiver
The face of the 22 GHz feed horn (below the clamping ring) is 1442 mm above the R1 (inner) targets.
Subre ector
- The subre ector's axis is parallel to the main re ector axis to within 0.5 arcmin (NS) and 2.24
arcmin (EW).
- The subre ector's axis is oset from the optical axis 1mm towards the North, and 5mm towards the
West.
The Targets
radius(mm) height above vertex (mm)
R1 (inner) 1867 130.7
R5 (outer) 9335 3145.8
R6 (outer) 10935 4277.6
The lower surface of the optical-axis locating jig is 1511 mm above the vertex.
4.4 MOPRA
The mopra antenna was checked in January, 2000. The C/X receiver was found to be within 2 mm
of the optical axis; the mm receiver was 10mm oset. Since a repositioning of the index pin locator
could not reduce this error, we moved the receiver. The error is now less than 3 mm.
Subre ector
- The subre ector's axis is parallel to the main re ector axis to within 0.75 arcmin in both directions.
- The subre ector was adjusted laterally, 6.5 mm in the NE direction.
5 This Data and the October, 1999 Holography
The holography surveys of the Narrabri antennas made in October, 1999 measured the subre ector
positions; these are in reasonable agreement with the alignment observations. We can combine the
receiver and subre ector measurements to predict the osets that would be seen by the holography.
antenna EW oset (mm) NS oset (mm) predicted EW predicted NS
CA01 +0.7 -0.1
CA02 +0.7 +0.0 +1.0 +1.0
CA03 -2.5 +1.2 (-1.8) 0.
CA04 -3.9 +0.5 -4.0 0.
CA05 +1.5 -2.5
Table 4: The subre ector positioning, measured by holography (rst two columns) and predicted from
the survey data
The holography receiver (at 12.75 GHz) was installed over the C/X frame, so should share the feed
osets of that receiver. The results for CA03 are unclear, as the C/X turret location has been adjusted
twice since the holography observations.
6

6 Discussion and Recommendations
 All the receivers measured were installed with more than adequate precision to ensure that the
antenna gain was never compromised. You need (
=)  7 to see a 1dB reduction in gain.
The new 80 GHz receivers are just at the point where higher precision is relevant.
We expect to make the nal sub-re ector adjustments on the basis of holography measurements.
Since the error signatures of feed and sub-re ector osets are so similar, we should ensure that
the holography and mm receivers have the same oset - and preferably zero oset.
 The feed osets lead to a pointing error, described as the squint term in the pointing model. It
amounts to 3.6 arcsec/mm at all frequencies. This means that we need receiver-specic pointing
model corrections, as we need to be able to transfer pointing models from one frequency to
another.
|||||||||||||||||||||
le:/nfs/corfu0/survey/txt/nbr report.tex
7