Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.arcetri.astro.it/~olmi/LMT/STACVIII_abs.pdf Дата изменения: Thu Nov 3 02:18:08 2005 Дата индексирования: Tue Oct 2 10:21:19 2012 Кодировка: Поисковые слова: lightning

WFS, cross-polarization, seeing-monitor

1

WAVE FRONT SENSOR, CROSS-POLARIZATION AND SEEING MONITOR: SUMMARY OF ACTIVITY
L. OLMI(1) , D.J. HOPPE(2) and D. HIRIART(3)
(1) (2) (3)

LMT/GTM Project, olmi@lmtgtm.org JPL, Daniel.J.Hoppe@jpl.nasa.gov UNAM (Mexico), hiriart@bufadora.astrosen.unam.mx

1
1.1

WAVE FRONT SENSOR
intro duction

1

The LMT/GTM pointing accuracy will ultimately be limited by tropospheric turbulence, and to achieve theoretical pointing specs and beam shape over a wide range of wavelengths and ambient conditions, the radio seeing must be measured and compensated. The FCRAO Anomalous Refraction data show in particular that most (> 70%) of the events last less than about 5 sec (Olmi 2000, in preparation), and they will thus strongly affect OTF mapping, which will become the most common observing mode on the LMT/GTM. In the past we have proposed a "Wave Front Sensor" (WFS) that would: (1) sense the wave front tilt, and (2) compensate in (quasi) real-time for the phase gradient across the antenna aperture by tilting the subreflector. The WFS is based on a single, 3-channel scanning 183 GHz radiometer, mounted slightly off-axis. A pick-off mirror (POM) located on one side of M3 collects the atmospheric emission, which is then reflected onto another flat mirror or "tipping flat" (TF), to steer the beam at the Cassegrain focus, and is finally focussed onto the WFS after a focal ratio conversion (see Fig. 1). In a single scanned radiometer the absolute gain calibration is not critical because a differential measurement is involved. However, the scanned radiometer has two main problems that needed a detailed analysis: (i) because the WFS is positioned off-axis, the spillover is different when the beam is scanned across the antenna aperture, and may exceed the required sensitivity; (ii) the illumination of the primary affects also the level of overlapping of the beams along the line of sight, which may limit the ability of the WFS to comp ensate the phase gradient. We analyse these two issues in the next section.

1.2

Spillover

Figures 2 and 3 show the expected incremental sky brightness temperature as a function of frequency when the water vapor pressure (PWV) is 1 and 4 mm, respectively. One can see that when PWV=4 mm the incremental Tb can drop below 100 mK in ch. 2 and 3, whereas ch. 1 becomes insensitive. This means that the WFS must be able to measure brightness temperature differences with a sensitivity better or much better than about 100 mK. To calculate the spillover we first designed the optical layout of the WFS using CODE V and then imported this model into a Physical Optics (PO) program. The PO code was run on a 128 node parallel processor in batch mode in order to obtain accurate results for the
1

This section by L. Olmi and D.J. Hoppe


WFS, cross-polarization, seeing-monitor

2

TERTIARY MIRROR

PICK-OFF MIRROR

TO SUBREFLECTOR

WFS

CASSEGRAIN FOCUS

Figure 1: Top view of a section of the LMT/GTM receiver room, showing the cross-sections of the tertiary and pick-off mirrors when the telescope is pointed at the horizon and at the zenith. The cross-sections of the Gaussian beams (at = 3 mm) along the chief ray and at the edges of the main field of view (telescope at horizon) are shown. The pick-off mirror is sitting slightly to one side of the tertiary mirror and the Gaussian beam at = 1.64 mm is also shown. large structure at 183 GHz. In the current design the telescope and the reimaging mirror, which converts the Cassegrain focal ratio to Frx = 29.2 at the feed-horn, form a Gaussian Beam Telescope. The POM is 81.6 cm distant from M3 (or 5 .3 on the sky) and 56 cm in diameter. The total motion of the TF is only ±1 .3 and can thus be controlled fairly easy with fast ( 1 - 2 Hz) actuators. A 30 dB gain corrugated feedhorn was designed and used along with a re-imaging ellipse to illuminate the TF for the PO calculations. The beam-waist on the antenna aperture is 2w = 9 m. This model is used in the PO program to calculate the fields over all optical apertures and thus the individual contribution to spillover for each position of the TF can be obtained The top panel of Fig. 4 shows the distribution of the surface currents, on both subreflector and main dish, at one of the four sampling positions on the antenna aperture. The center of the beam-waist is at a distance of 11 m from the antenna center. The overall spillover is shown in the bottom panel of Fig. 4 where one can see that the differential spillover among the four sampling positions is < 25 mK, small enough to detect the variations in Tb .

1.3

Tip-tilt correction

Because the "footprint" of the beam on the antenna aperture in each of the four positions is very close to the antenna center and the optical axis, as shown in Fig. 4, it was necessary to verify that the sampling of the antenna aperture would be able to sense and then compensate the phase gradient across the incident wave front.


WFS, cross-polarization, seeing-monitor

3

Figure 2: Incremental Tb for an average pwv of < w >= 1 mm at 4600 m, i.e. for very dry conditions. Tamb = 270 K. Tsys (z = 0; = 230 GHz) = 213 K, (z = 0; = 230 GHz) = 0.058 (assuming Trec = 60 K). The solid and dashed lines correspond to wrms = 0.01 and 0.005 mm, respectively. The three frequency bands of the radiometer are shown as vertical dashed lines (in the USB only). The first LO is assumed to be at 183.3 GHz. We thus run several simulations and verified that, in most cases, even with beams as close as 10 m to the antenna center is possible to partially or totally compensate the AR pointing error. The example in Fig. 5 shows the phase screen over the antenna aperture before and after the tip-tilt correction has been applied, and one can see that most of the phase gradient has been removed from the wave front.

2

CROSS POLARIZATION ANALYSIS2

The cross-polarization analysis of the optics in the LMT/GTM had mainly two goals: (i) determine the cross-pol characteristics of several alternative optical designs for SEQUOIA; and (ii) design a CODE V procedure that would yield comparable results to those of a much more sophisticated and computer-intensive PO program. To calculate the cross-pol the PO program assumes a linearly polarized square corrugated feed-horn (a single SEQUOIA pixel) at a given position in the receiver focal plane and then propagates the field through the reimaging and telescope optics. The cross-pol is calculated comparing the X - and Y -components of the far field. The result is shown in Fig. 6 for several optical designs and various positions in the focal plane or, equivalently, on the sky. In CODE V, on the other hand, one must specify the polarization state of the input beam (i.e. the source on the sky) and the program then models the effects of each interface or optical surface on the polarization state of the traced rays. To calculate the resulting cross-pol at the focal plane we had to carefully write a CODE V procedure to overcome several technical difficulties.
2

This section by L. Olmi and D.J. Hoppe


WFS, cross-polarization, seeing-monitor

4

Figure 3: Same as Fig. 2 for 4 mm pwv, i.e. the median conditions at the LMT/GTM site. Tsys (z = 0; = 230 GHz) = 363 K, (z = 0; = 230 GHz) = 0.22. Table 1: Values of the cross-pol component in dB relative to the co-pol component, obtained through PO and CODE V for the gsoff45 2 design also listed in Fig. 6. Position in FOV (x, y ) (2 , 0) (0, 2 )
(a)

X -polarized field(a) PO CODE V -42.8 -42.1 -27.6 -27.4

Y -polarized field(a) PO CODE V -42.4 -19.9 -27.6 -17.9

Polarization of feed-horn for PO and input beam polarization for CODE V.

In Tab. 1 we compare the results obtained with the PO program and CODE V in one specific optical design. Although we get a very good agreement between the two methods when the feed is X -polarized, the results are clearly different when a Y -polarized feed is considered. This disagreement may be due to the fact the any phase coupling has been neglected in CODE V so far, and we are currently working on this problem.

3

SEEING MONITOR

3

The radio seeing monitor has been recently upgraded to a "hybrid" design, which combines the concepts used in similar instruments developed by Radford et al. (1996) and by Lay (1998): the front end is a classic heterodyne phase locked receiver while the back end uses an IQ demodulator scheme to measure the phase difference between the two sampled paths. It has the advantages of using a single down conversion stage and a direct estimate of the phase difference.
3

This section by L. Olmi and D. Hiriart


WFS, cross-polarization, seeing-monitor

5

The signal source is the monochromatic tone beacon broadcasted by the Mexican communication satellite Solidaridad II at a frequency of 11.715 GHz. The beacon has a stability better than 0.3 ppm and it has vertical polarization mode. The position of the satellite is at 113 W longitude on the ring of geostationary orbits, so the antennas located at Cerro la Negra are oriented at an azimuth of 222 towards the east from the true north and at elevation of 61 . The beacon signal having a 3 MHz bandwidth is received by two commercial 80cm satellite dishes. The maximum baseline of the radio seeing interferometer is about 50 m. The antennas are mounted on metallic posts and attached to a solid foundation; they were reinforced with a back structure to increase their resistance to wind deformation. Also, since the receiver feeds and the beacon satellite are linearly polarized, a method to finely rotate the receiver feed was implemented to receive the maximum strength signal. Fig. 7 presents a block diagram of the implemented system. The interferometer uses a single stage of down conversion. Low noise direct TV broadcasting receivers amplify the Ku band signal. These commercial receivers were modified to provide the amplified signal to an external down conversion stage were the signal is converted to an IF of 880 MHz. The down conversion stage accepts the external LO signal generated by a coaxial resonator oscillator phase locked to a common temperature stabilized crystal oscillator. The reference frequency of this crystal oscillator is 89.5 MHz. The signals from the local oscillator reference and the received IF signals are transmitted from the antennas to the processing electronics through thermally stable cables to reduce the phase drift of the instrument. Cables are buried underground or thermally isolated for external runs. Different cables are used to send the IF and the reference oscillator signal. The IF signal is passed through a 10 MHz bandwidth passband crystal filter centered at a frequency of 880 MHz. This allows to increase the signal to noise ratio when the single monochromatic sinusoidal signal is sent to the IQ demodulator. Finally, the IQ demodulator delivers the in phase and quadrature components of the two signals to obtain the phase difference. These signals are read by a 16 bit A/D converter at a sampling rate of 10 sample per second and sent to the main computer for further processing. The site preparation at CLN has been practically completed, including: · 60m в 20m flat platform · extended lightning protection · optical seeing monitor (by E. Carrasco) · access road · instruments shelter · power The construction of the antennas and related hardware has been completed. Some of the revised control electronics and data acquisition system must still be completed and tested.


WFS, cross-polarization, seeing-monitor

6

Tipping Mirror at 136.20 deg. [Left] R=11.06 m

Subreflector

Main Reflector

WFS Noise Temperature Calculations (Sub Spill->200K, Others->300K)
150.00 140.00 130.00 120.00 110.00 Noise Temperature (mK) 100.00 90.00 80.00 70.00 60.00 50.00 40.00 30.00 20.00 10.00 0.00 135.02 deg. 136.2 deg. 135.62/91.2 deg. Tipping Mirror Rela tive Motion main sub pick-off tip

Figure 4: Top. Surface currents on the subreflector and main reflector for one of the four beam positions of the WFS. Currents are normalized to their peak, and a range of 28 dB is plotted. Bottom. Spillover contributions for the four beam positions: left, right and up/down.


WFS, cross-polarization, seeing-monitor

7

Figure 5: Top. Single time-realization of the incident wave front over the antenna aperture. Bottom. Corrected wave front obtained by removing the linear phase gradient across the aperture using the parameters described in the text.


LMT Re-Imaging Optics: Cross Polarization Performance

60.00

X Polarization, 2-Arc Minc X Scan Y Polarization, 2-Arc Min Y Scan X Polarization, 2-Arc Min Y Scan

Y Polarization, 2-Arc Min X Scan

50.00

40.00

dB Relative to Co-Pol Peak

30.00

WFS, cross-polarization, seeing-monitor

20.00

10.00

0.00 cas_focus gsoff45_2 gsoff22_2 System gausoff4_3 gsoff23_2 gsoff44_2

8

Figure 6: Cross-polarization performance of various LMT/GTM optical designs obtained using PO. The values in Tab 1 correspond to gsoff45 2.


WFS, cross-polarization, seeing-monitor

9

Figure 7: Block diagram of the hybrid radio seeing monitor.