FCRAO Cookbook for Beam Switching Observations
S. Leon and M. Yun (February 2003)
This a short cookbook for conducting
"beam switching" observations using the 4x4 pixels
SEQUOIA array on the FCRAO 14-m telescope.
Any arbitrary pair of pixels can be selected for a beam switching
observation, but generally it makes a good sense to choose two
adjacent pixels in AZ direction. In the example observation
described below,
pixels 11 and 27 are assumed to be the ON-SOURCE or the (0,0) position.
The OFF position is taken to be the next adjacent pixels in azimuth,
i.e. the pixels 7 and 23. The ON and OFF positions are beam-switched
so that the source is always observed with either the ON or the OFF
beam in both polarizations.
The SEQUOIA receivers have a very large instantaneous
bandwidth (16 GHz) and tuning is done on the IF with the backend.
So tuning is not required other than by specifying the line+sideband
with the CONF command, and there is virtually no overhead for tuning.
Getting Started
Connect to the FCRAO computer by typing:
> ssh -l observer fcrao.astro.umass.edu
Contact Mark Heyer (heyer@astro.umass.edu) or one of the postdocs
for the password. Once logged in, start a terminal emulator by typing
> mterm
You are now connected to the MODCOMP running the telescope.
After providing some contact information, look for ?NET
prompt. (You will also need another password in order to take
command of the control program.) To refresh the screen, type
?NET REFR
?COM POINT
Update Telescope Pointing
Telescope pointing is most commonly done by observing SiO masers.
Locate an appropriate source from the
pointing source catalog.
In this example, we are using RLEO.
MAIN NAME RLEO
IFTYPE 2
STAR QTUNE
CONF RESET
SAME REF AZOFF -.2
CONF 1 COR LINE SIO-4 LSB
TMAI 10 TREF 10 RPT 1
CBAND 50
APMODE 2
REF STAR CAL STAR APOINT
Pointing is stable over many hours. An update is
recommended mainly after the sunrise and sunset. APOINT
routine goes through one set of 5-point measurements, then
the focus optimization, followed by another 5-point measurement.
It takes a total of 10-15 minutes, depending on the convergence
of the pointing measurements. To inspect the pointing progress,
type SCRE 19. (SCRE 1 to return to the main screen)
Pointing offsets are automatically adopted every time the 5-point
pointing measurement converges.
Beam Switching Observation using Position Switching Mode
The example shown here is CO (1-0) observations of NGC 2798 using
the QEF backend. For observations of Galactic sources, change the
spectrometer setting to the digital correlator accordingly.
MAIN NAME NGC2798 RA 091722.8 DEC 415958.0 VLSR 1726
IFTYPE 1
CONF RESET
CONF 1 QEF LINE CO-1 USB
QSELECT 11
START QTUNE
TMAI 18 TREF 18 RPT 10
ABC = SAME REF AZOFF -.0248 START CAL START PS REPE 5
ABC
Each integration consists of 10 repeats of 18 seconds ON source
and 18 seconds OFF source, with a total of 360 seconds.
SAME REF AZOFF -.0248 defines the OFF beam to be the
next pixel in azimuth, -0.0248 degree (89") away. START CAL START
PS REPE 5 initiates 5 repeats of CAL (Tsys measurement) and
PS mode integrations. One can define a new function as in
"ABC = SAME REF..." and use this fuction call rather than typing
long commands every time.
If you want to interrupt an integration, type
KEYBOARD
STOP PS (if you are doing a position switch PS)
Stopping the Observing Session
DROP (puts the telescope in the stow position)
HALT (Halts the telescope)
EXIT
For more information look at :
http://donald.phast.umass.edu/~fcrao/library/manuals/
Data Reduction
If you do not know exactly what are the correct scan numbers,
here is a short direction from Mark Heyer
on how to do this usin SPA.
Signal from the target source is found in two ON pixels (11 and 27)
and in two OFF pixels (7 and 23). Ultimately we want to sum all
data from these four pixels together to obtain the best S/N spectrum.
However, all OFF position pixels (7 and 23) have inverted
spectra because the target source is in their OFF position.
Furthermore, these OFF spectra cannot be summed together in a
straight forward way because sky rotation causes the coordinate
information on the OFF pixels to change with time.
To reduce the data in CLASS, you should:
- invert all OFF spectra; and
- relax the coordinate match parameter to ignore differences in
the header coordinate information (e.g. set match 200)
There is a macro called qef.fcrao which
can be run within the FCRAO package (which is a local version of
CLASS running on the fcrao computer). To start, type
>fcrao
fcrao> @qef first_scan last_scan v1 v2
This will dump all the scan in a directory and get the ON pixels (11+27) and
the OFF pixels (7+23) corrected. The "v1" and "v2" are to set the
line window for subtracting a DC offset and recording "rms" value for
each spectrum. Then do the usual reduction in CLASS:
fcrao> sum
fcrao> set win v1 v2
fcrao> base 1
fcrao> gauss
fcrao> fit
fcrao> write
The output is a file called "temp.14m", which should be renamed to be
saved.
Last modified by M. Yun (12/3/2003).