|
|
|
|
Tracking, Pointing, and Focus
4.1 Tracking CapabilitiesThe 12 Meter Telescope can track both stationary (sidereal) sources or fast moving sources such as the Sun, Moon, planets, comets, or satellites (called “ephemeris objects"). The positions of stationary sources may be entered into source catalogs by the observer or directly into the on-line system by the operator. Stationary source positions may be given as
An example of the 12m catalog format is the following:
where column 1 = right ascension (HH:MM:SS.S) or galactic longitude (DDD.DDDDD), column 2 = declination (DD:MM:SS.S) or galactic latitude (DD.DDDDD), column 3 = coordinate epoch (B1950, J2000, APPARENT, or GALACTIC), column 4 = source name, column 5 = source velocity, column 6 = source velocity reference frame (LSR, HEL, GEO, or TOP), where LSR = local standard of rest velocity reference frame, HEL = heliocentric velocity reference frame, GEO = geocentric velocity reference frame, and TOP = topocentric velocity reference frame, column 7 = source velocity type (RAD, OPT, or REL), where RAD = radial velocity type, OPT = optical velocity type, and REL = relativistic velocity type
The sky frequency equations corresponding to the above velocity types are given in Equations 5.4.
4.1.1 Ephemeris ObjectsTo track ephemeris objects (comets, asteroids, etc.), the control system requires three positions, centered upon the present time. The tracking algorithm then makes a cubic spline interpolation to determine the source position at the specific time of observation. The observer must prepare a catalog in the correct format in order to observe ephemeris objects other than the planets and the Moon. The basic format needed by the control system is show below.
where the first three lines are descriptors for the object name, velocity frame, and velocity type. The next three lines are comments. Each column thereafter is defined as follows: D 45.7745 = the topocentric right ascension in degrees. D 41.3975 = the topocentric declination in degrees. NOTE: You can also use a geocentric right ascension and declination in the last two locations, but if you do you must enter the correct horizontal parallax (see below). J 50182.000720 = the Modified Julian Date (mjd) of this record in the TT (Terrestrial Time, which used to be called Terrestrial Dynamical Time, or TDT) system. Note that TT = UTC + 63.184 sec after June 30, 1997. 0.0 = the horizontal parallax (the angle from you to the object to the center of the earth in arc seconds). The horizontal parallax is zero in this case since we are using topocentric coordinates. If one uses geocentric coordinates, the horizontal parallax must be non-zero and is given by:
HP = arcsin
55.84 = the topocentric velocity. Non-planets default to topocentric velocities. Planets default to geocentric positions and velocities. After you create the ephemeris put it in a file with the extension eph in the observers home directory - /home/obs/ini/comet.eph for example. If you do not have an ephemeris generation program, we have a copy of the “TBEG4" two body ephemeris generator program written by Don Yeomans which you can use. This program will produce a 12m-format ephemeris file given a set of orbital elements. Contact Tom Folkers if you would like to use this program.
4.2 Tracking Limits4.2.1 Elevation LimitsMechanical limits constrain the movement of the telescope on both the azimuth and elevation axes. Under normal source-tracking operation, the control system software prevents the telescope from moving into these limits. The telescope is also equipped with fail-safe hardware limit switches that will turn the drive motors off and apply the brakes before the telescope can be damaged. When the telescope reaches a lower elevation of 15.0º, it begins to depress a safety spring on the elevation stop. Tracking can continue until the elevation drive motor begins drawing excessive current; this usually occurs at an elevation near 14.5º. A final limit switch is tripped at 13.8º elevation. The control computer will allow a source to be tracked up to an elevation of 90º. The pointing equations diverge at the zenith, however, and elevations greater than 88º should be strictly avoided during observations. For routine observations, we recommend that you avoid elevations >80º if possible, since both tracking and pointing degrade in that range. For maintenance purposes only, the telescope can be driven to an elevation of 92º before tripping a final limit switch.
4.2.2 Azimuth LimitsTo prevent the over-wrapping of cables, the telescope has hardware and software azimuth limits. The control system will not allow the telescope to rotate through 66.8º azimuth to acquire or track a source. As they rise, sources with declinations between 27º and 38.75º will pass through 66.8º azimuth above 15º elevation. When a rising source reaches an azimuth of 66.8º, the telescope will rotate 360º to re-acquire the source. The region of the sky where azimuth transitions occur is displayed in Figure 4.1. If an integration is in progress when the transition azimuth is reached, the integration is halted while the telescope rotates around to re-acquire the source. The integration is resumed when the source is reached again. If it is undesirable for this to happen, you should cease the integration before reaching the transition azimuth.
4.3 Tracking Error ToleranceThe 12 Meter Telescope is vulnerable to tracking errors produced by wind gusts. The tracking software has a provision for rejecting from the integration any data sample that was taken with the telescope off source. The basic timing cycle of the 12m control system is 100 milliseconds (0.1 sec). After each 100 ms interval, the system checks for error conditions such as tracking errors or a loss of frequency phase lock. If an error condition is detected, the data sample collected during the last 100 ms interval is not averaged with the rest of the data. You can choose how much tracking error you are willing to accept. This number is called the “tracking tolerance" and is displayed on the on-line status monitor in the lower left-hand corner under the heading “TOL". The 12m typically experiences 3-4” tracking errors under calm conditions. The tolerance is usually not set to less than 5” for this reason. Typical choices for the tolerance are 5” for 1.3 mm and shorter wavelength observations, and 10” for 2 and 3 mm observations. Figure 4.2 shows the amount of signal that is lost by a given pointing error, expressed as fractions of a FWHP beamwidth.
4.4 Sequence of Position Computation OperationsThe sequence of computer operations that is executed when seeking and then tracking a source is as follows:
Figure 4.2: Loss of signal due to pointing error.
4.5 Subreflector Beam ThrowThe 12 Meter Telescope is equipped with a nutating (chopping) subreflector that is used for beam-switched observations in both the spectral line and continuum modes. Other observing modes, such as position switching, are made with the subreflector locked in place. Obviously, the throw position of the subreflector, whether it is chopping or locked in place, affects the pointing of the telescope. The current 12m subreflector switches only in azimuth, although mount misalignments may produce a small component in the elevation direction (usually these are less than 1-2”). The pointing equations for the 12m are set up for 0 offset in subreflector position, i.e. with the subreflector axis aligned with the electrical axis of the primary reflector. When the subreflector is offset so that the telescope must be moved in the positive azimuth direction to bring the source into the beam, the subreflector is said to be in the “+BEAM”. When the telescope must be moved in the negative azimuth direction to acquire the source, at least with respect to the +BEAM, the subreflector is in the “-BEAM”. Figure 4.3 illustrates this convention. Both the spectral line and continuum backends are configured so that the +BEAM signal is positive and the -BEAM signal negative.
Figure 4.3: Subreflector +Beam/Beam conventions.
The control system software takes account of the subreflector throw setting when positioning the telescope. The computer does not read the subreflector setting automatically, however; it depends upon the operator to manually set the throw value. The subreflector is physically set by turning two dials on the subreflector control chassis. Default values of subreflector throw are stored in the computer: for example, ±2′ for 3 mm and ±1′ for 1 mm. You can check the beam throw by making a continuum cross-scan on a strong source. Continuum mapping rows are explained in detail in Chapter 6.
4.6 PointingYou are responsible for determining the residual azimuth and elevation pointing offsets, relative to the nominal telescope pointing. These offsets are usually of the order 10″- 20″ and often vary across the sky. In particular, the offsets often show an elevation dependence. The pointing of the 12m may drift over a period of several months by 10″ or so, and you are cautioned not to assume that old pointing data are still valid. If a project requires accurate pointing, observers should budget 5 - 10% of their total observing time to pointing checks. High frequency projects, for which the FWHP of the main beam is 30″ or less, obviously require more attention to the pointing. You may perform pointing checks in either continuum or spectral line mode. Pointing sources must have an angular brightness distribution that is compact compared to the antenna beam. In addition, they must have well-determined positions and must be strong enough to point on in a reasonable amount of time, e.g. 10 minutes or less. The best pointing sources are the major planets Venus, Mars, Jupiter, Saturn, and Uranus, observed in continuum mode. These sources have strong flux densities throughout the millimeter band, although Uranus is faint at 3mm and longer wavelengths. To obtain complete sky coverage, particularly north of the zenith, the planets must be supplemented with other pointing sources. A tabulation of these sources and approximate flux densities at 3 and 1 mm is given in Table 4.1. This table includes the sources traditionally used at the 12m together with sources found useful at the IRAM 30m telescope. The sources with no listed flux density are on the order of 1 Jy in strength. It should be noted that the flux densities of many of the extragalactic sources in the table can be highly time-variable. For an up-to-date listing of measured 3mm fluxes, see http://dopey.haystack.edu/cmva/quasar.list. The spectral line emission from a number of sources can be used for pointing purposes. Emission from rotational transitions of CO, HCN, and SiO are often strong enough and angularly compact enough to provide good pointing results. A list of spectral line pointing sources and their strong molecular emission lines are listed in Table 4.2.
4.6.1 Continuum and Spectral Line Five-Point MeasurementsThe five-point mapping
option is the standard observing technique for determining the antenna pointing
of the 12 Meter Telescope. Continuum five-point measurements can also be used to
measure the flux density of a strong source with the highest precision, or the
flux density of a source whose position is not exactly known. The five-point map
consists of five successive ON/OFF
sequences (for continuum and beam-switched spectral line measurements), or five
successive ON and OFF
measurements (for position-switched spectral line measurements), or five
successive ON measurements with one
associated OFF measurement (for
total power spectral line measurements). The five sequences or ON
measurements are positioned in azimuth and elevation relative to the nominal
source position, as shown in Figure 4.4. The numbers identifying the positions
give the time order in which the sequences are taken. The offset for each
position (“HP” in Figure 4.4), is a selected offset (in arcsec) that is usually
chosen to be close to half the HPBW of the telescope at the observing frequency.
HP should be somewhat larger if the source is extended; for a planet that is
resolved by the beam, chose HP to be approximately the semi-diameter of the
source. Table 4.1: Galactic and Extragalactic Continuum Pointing Sources
To make a five-point map proceed as follows:
Continuum Five-Point:
Spectral Line Five-Point:
Table 4.2: Spectral Line Pointing Sources
Figure 4.4: Five point mapping scan sequence. 4.6.1.1 Continuum Five-Point AnalysisContinuum five-point measurements are processed automatically by the on-line dataserver. The results from these on-line fits are sent automatically to the operator for entry into the control system. To process a continuum five-point measurement manually, use the condar resident procedure five, which can be abbreviated as f. You can look at the results for each channel separately by using the command Condar > scan number f Channel 1 is selected by specifying subscan number .01, while channel 2 is selected with subscan .02. The display (Figure 4.5) provides the observer with the best-fit positional offsets in azimuth and elevation, along with the estimated peak flux density of the source, both for a constrained fit of Gaussian half-width HP and an unconstrained solution. The optimum pointing offsets (labeled CORRECTED OFFSETS) are also printed, assuming the given position of the source to be correct.
4.6.1.2 Spectral Line Five-Point AnalysisThe display of the map and the fit for pointing offsets is somewhat more involved than for a continuum five point. There is a standard line procedure which processes all types of spectral line five-point measurements. To process a measurement, within line type: Line> center scan number fivel1 (for channel 1) Line> center scan number fivel2 (for channel 2) You will be asked some questions by the routine:
AVERAGE THE TWO RECEIVERS? (1=YES, 0=NO) - This question is asked only if the data were taken in series mode. If the question is answered “yes”, the routine averages the first and second filter bank to improve sensitivity. You should answer “yes” only if you are reducing the first filter bank; you will usually not need to reduce the two filter banks separately if averaging the two. If the data were taken in parallel mode, the two polarization channels in each filter bank are automatically averaged.
Figure 4.5: Continuum Five-Point Analysis Example
Figure 4.6: Display of each scan produced by the line procedure fivel.
USE OLD BASELINE & INTEG. REGIONS? (1=YES, 0=NO) - The observer must define baseline fitting and spectral line integration regions for this routine. If these regions have already been accurately set, as, for example, in a five-point of the same source performed immediately prior to the present one, you can answer “yes” and speed up the reduction process considerably. If you answer “no”, the spectrum at the center of the map is displayed on the screen, and the question ENTER # OF BL SEGS is printed. This part of the f routine is the same as the bset procedure. The observer must enter the number of baseline segments to be used to fit a least-squares baseline. [NOTE: The order of the fit is determined by whatever nfit is set to (the default is 1 - see the line data reduction manual for a description of the baseline fitting facilities).] When the number of segments to be fit is entered, the graphics crosshairs appear and you must mark off the two ends of each segment with the crosshairs (move the crosshairs to the desired position and strike any mouse button). When this operation is finished, the baseline is removed from the center spectrum and the spectrum is replotted. The prompt ENTER INTEG. REGIONS is printed and the crosshairs reappear. You should then mark off the two velocity/frequency extremities of the spectral line (or whatever region you wish to integrate over). DISPLAY EACH SCAN? (1=YES, 0=NO) - If you answer “no”, the map and the fit for new offsets are printed out. If you answer “yes” each scan of the map, with baselines removed, is displayed on a single plot, one spectrum on top of the other (see Figure 4.6). TYPE 1 TO CONTINUE OR 0 TO STOP - If you approve of the fits, answer 1 and the map and fit for new offsets will be displayed. An example of the five-point map with explanatory annotations is given in Figure 4.7.
4.6.2 Pointing Model EquationsIn addition to coordinate conversions, the telescope tracking program incorporates pointing corrections. The pointing corrections take into account resolver zero offsets, differences between electrical and mechanical axes, axis tilt, etc. The coefficients of the pointing model equations are determined during dedicated optical and radio pointing source measurements several times during the observing season. The pointing models used for the 12m are described in detail in Appendix A.
4.6.3 Pointing Data Analysis ProgramEach time a continuum five-point measurement is made, the on-line data server performs a fit to the results and writes them to a permanent file for later analysis. A program called gpoint is available on any of the mountain workstations for display and analysis of the recent pointing history of the telescope. To start the program, type gpoint from any login prompt. gpoint is menu-driven, so you need only select the appropriate entries in each menu to configure the data selection and display parameters to suit your needs. An example of a gpoint plot is given in Figure 4.8.
4.7 Focus4.7.1 Axial FocusTo bring the incident radiation to a focus at the receiver feed horn, the subreflector of the 12m may be moved in and out along the electrical axis of the telescope. The focus exhibits an elevation dependence which is automatically corrected by the control system computer, according to the equation F(E) = F0 + 2:8 sin(E) (4.1) where F(E) is the focus setting in millimeters, E is the elevation, and F0, the subreflector focus setting at 0º elevation, is to be determined by the observer. F0 is given in millimeters, and larger values of F0 represent greater distances between the dish surface and the subreflector.
An automatic procedure called a focalize (this word really is in the
dictionary!) will determine the best value for F0. An analysis
routine of the same name exists in condar and line for displaying
and fitting the results. Instructions for performing a focalize are given
in §4.7.2.
Figure 4.7: Spectral line five-point analysis example.
4.7.2 Determining the Axial FocusThe axial focus of the telescope should be checked occasionally during an observing run. It is often most convenient to make a focus check after a continuum five-point pointing measurement since both measurements use the same kinds of sources (strong continuum point sources). It is particularly important to check the focus after nightfall and daybreak. Focusing is performed via the focalize procedure. This steps the radial distance of the subreflector from the telescope vertex through seven positions centered on the nominal best setting (F0). Larger values of F0 represent greater distances between the dish surface and the subreflector. The incremental movement between positions (WL) is set to half of the observing wavelength by default, but can be set by the observer. Perform a focalize as follows:
Focalize measurements are processed automatically by the on-line dataserver, the results of which are sent automatically to the operator for entry into the control system. To reduce the focalize in condar use the commands, Condar > scan number focalize where each channel is selected by specifying either subscan .01 for channel 1 or .02 for channel 2. The output from the focalize procedure is shown in Figure 4.9. The fit should be smooth and “quasi-Gaussian". (If not, repeat the focalize. Increase the value of WL if you think this is necessary.) The display will give a best-fit value for F0. Where appropriate, you should average the values for the two channels.
4.7.3 Lateral FocusThe position of the subreflector for optimum gain shifts in the elevation direction, with changing elevation angle. The change in antenna gain produced by this effect is not significant at 3mm wavelengths but could be as much as 20% or more at 1.3mm and shorter wavelengths. The 12m is equipped with lateral “north-south” and “east-west” translation stages at the prime focus to eliminate this loss in antenna gain. The computer is able to automatically control the positioning of these stages, simultaneously applying an appropriate correction to the telescope pointing in elevation and azimuth, respectively. This correction to elevation pointing is 31.5″/mm (which is the prime focus “plate scale” derived in Appendix G) of movement in the north-south focus or east-west focus stages. In the future, the coefficients necessary to control the movement of the north-south and east-west translation stages will be installed in the computer and applied automatically. At this time, the north-south and east-west translation stages can be set to a given position, but are usually left fixed in this position. Automatic procedures for fitting for maximum north-south or east-west gain, called nsfocal and ewfocal are available. These routines operate analogously to the focalize procedure. Normally, only staff astronomers use these procedures; we do not recommend that observers spend time trying to optimize this further. Figure 4.9:
Focalize display example. |
|
Copyright Arizona Radio Observatory. |
