Berkeley Absolute Flux Calibration System (Gibson/Welch)
Last Update: JeffMangum - 12 June 2009
Contents
Berkeley Absolute Flux Calibration System (Gibson/Welch)
A draft report describing this system has been produced by Jim Gibson. Comments to this draft should be posted here...
Jack Welch has written a nice summary description of the Berkeley Absolute Flux Calibration System.
Note from Jim Gibson summarizing recent 86 GHz measurements
In July and August we (Jim Gibson, Jack Welch et al) calibrated the gain of
one of the BIMA 6m antennas at Hat Creek at 86 GHz by comparison to a (small!)
standard gain horn.
This comparison was done using the interferometer with the four antennas
available there at the time. A bright source (Venus, mostly) is observed
with the test antenna vs the others and again with the standard horn vs the
others. That is, the standard horn becomes an element of a baseline. The
ratio of fringe amplitudes is the voltage gain difference.
Having done this, we used the calibrated antenna to measure the flux of venus
and Jupiter at this frequency, referencing the signal to known hot and cold
loads on the receiver.
For this we built a special room temperature receiver with a waveguide
transfer switch to switch the input to the detector quickly between the main
dish signal and the standard horn signal.
A lot of data was collected and a lot of corrections and calibrations have yet
to be applied.
Since waveguide transfer switches are not going to be available for most of
the higher frequency bands for ALMA, Stephane Guilloteau proposed an
alternative calibration method using a standard gain horn used interfer-
ometrically, but without any transfer switch. This method is essentially to
use an amplitude closure relationship to establish the flux of a source
directly, knowing the gain of one antenna and the response of the electronics
below it. We took data on our measured sources using this technique as well
and will report a comparison of results and accuracies as soon as we sift
through all the debris.
A few notions are clear immediately after having made this attempt:
Pointing is a critical consideration for this sort of thing. For the 6m
BIMA dish at 86 GHz, pointing stability of about 5 arcseconds was desired
in order that pointing errors and deviations do not contribute significantly
to the final error. For higher frequencies, this tolerance will be tighter.
Every antenna involved in the interferometry must be well-pointed.
To know a source flux, the opacity of the atmosphere must be known. In our
more traditional method, the atmospheric opacity cancels out of the antenna
gain measurement and need only be measured at the moment when that calibrated
antenna is used to measure source flux. In the Guilloteau technique, the
source flux is measured gradually over hours (at least it was with our four
6m antennas on a very bright source) and so the opacity must be known
continuously and to good accuracy. Some sort of dedicated tipping device
or radiometric opacity measurer will need to be set up.
The cold load is a bit of a tricky business, filling the thing up with LN2
and warming it up again between usages. Some thought will need to go into
how to manage fairly continuous availability of a cold load signal, given
all the ways these loads can lose accuracy over time (icing up, dissolving
oxygen in the nitrogen, stress due to thermal cycling, etc).
A web page with more description, and pictures, and updates on the data
reduction will be available soon.
--
JeffMangum - 08 Sep 2004
Assigned to | Due date | Description | State | Notify | |
---|
JeffMangum | 2006-04-01 | Add new Gibson/Welch calibration system description to Calibration Plan | closed | JeffMangum | edit |
--
JeffMangum - 2009-06-12