[mmaimcal] Re: [Fwd: Re: CP beamsquint]

[Al Wootten]

Peter Napier writes:
 > Al,
 > >From Dick Crutcher's comments it looks as though we need a circular
 > polarizaion beam squint spec of something in the range 0.1% to 1.0% of
 > the primary beam. Do you concur?
Hi, Peter

I don't have very much experience with polarization so I defer to Dick.  I have
put aside a few minutes at the ASAC meeting to deal with remaining polarization
issues.  You may recall that the ASAC said, in their comments on the Wild/Payne
receiver specs:

* Section 3.2 POLARIZATION

    ASAC:  The ASAC report at
           http://www.cv.nrao.edu/~awootten/mmaimcal/asacreport/node12.html
           needs to be made more specific. Larry D'Addario and Steve Myers 
           have put work into this, and their draft recommendations are  
           located at:
           http://www.aoc.nrao.edu/~smyers/alma/polspecs-imcal.txt
           We invite comments on this document.

           The polarisation purity requirement mostly influences the optics 
           and horn/orthomode transducer. One would assume that the 1% goal 
           in calibration requires a cross-polarisation better than 20 dB, 
           and that all antennas should also be co-aligned in polarisation 
           to the same level. Polarisation experts should comment on this!
           Note that it may be difficult to reach this level with an 
           orthomode transducer.

Which sort of left things in the air.  I think this should be concluded at
the ASAC meeting, in addition to the question of beam squint.  Do you have
an opinion on what might be achievable with the ALMA antenna?  Dick thinks
0.1% may be difficult.  On an OH Zeeman experiment on the 43m, Barry relates
an experiment he did where beam squint of 0.27% was measured, and prevented
the success of the experiment (in collaboration with Heiles and Verschuur).
This agrees with what Dick says, and suggests a rather stringent requirement.

Al

 > -------- Original Message --------
 > Subject: Re: CP beamsquint
 > Date: Fri, 25 Aug 2000 15:21:49 -0500
 > From: Richard Crutcher <crutcher at astro.uiuc.edu>
 > Reply-To: crutcher at astro.uiuc.edu
 > To: Peter Napier <pnapier at cv3.cv.nrao.edu>
 > References: <39A6B48E.D5543EA9 at nrao.edu>
 > <39A6C28D.EA0AABB at astro.uiuc.edu> <39A6CC93.A29A655 at nrao.edu>
 > 
 > My experience with circular polarization observations is for Zeeman
 > observations
 > only, so I can only respond from that perspective. I've never seen an
 > antenna
 > with a beam squint as large as 10%, so I have no experience there. But
 > most
 > antennas seem to have a beam squint of about 1%. The IRAM 30-m antenna
 > had a
 > beam squint of 0.25 arcsec and a primary beam of 23 arcsec. This
 > definitely
 > introduced instrumental effects that caused significant problems. For
 > the IRAM
 > CN Zeeman observations (by the way, in the 1996 paper I cited we did not
 > achieve
 > a detection, but later we did and have published those results), the
 > experiment
 > could not have been done except for the special characteristics of CN
 > described
 > in the 1996 paper that allowed us to calibrate out beam squint effects.
 > For a
 > Zeeman experiment on a molecule without the hyperfine structure of CN
 > (SO is a
 > meaningful case), 1% beam squint would have been fatal. 0.1% would be
 > ok. For
 > synthesis observations beam squint is less crucial (see previous
 > discussion I
 > sent you), but I think for ALMA, single antenna data will be essential
 > for
 > inclusion with the interferometer. So we are back to the requirements
 > for single
 > antennas: 1% beam squint will compromise science, 0.1% will not (to the
 > best of
 > my knowledge). I've never seen an antenna with only 0.1% beam squint,
 > however;
 > that's a pretty tough spec to meet.
 > 
 > 
 > Peter Napier wrote:
 > > 
 > > Thanks Dick, I look forward to hearing if your student finds anything.
 > > 
 > > On the VLA the feeds are not displaced from the secondary focus. The
 > > feeds are on the focus, but the focus is displaced from the axis of the
 > > primary. You are correct that the formula for this case predicts the VLA
 > > beamsquint quite well. For ALMA one of the geometries under
 > > consideration is an on-axis focus but laterally displaced feeds. This
 > > case requires a different formula which is why I would like to check it
 > > on a known antenna.
 > > 
 > > Now here is a difficult question I would appreciate your opinion on. Any
 > > of the ALMA optics schemes that we are currently considering eg offset
 > > feeds or offset ellipsoid tertiary reflectors, are likely to have some
 > > residual beamsquint, although there are schemes to compensate and
 > > minimise the effect. The question is, how much residual beamsquint is
 > > tolerable? Does the total separation between the LCP and RCP beams,
 > > expressed as a percentage of the primary beamwidth, have to be <0.1% or
 > > <1.0% or <10% - pick one. Is this requirement specified anywhere that
 > > you know of for ALMA?
 > > 
 > > Peter
 > > 
 > > Richard Crutcher wrote:
 > > >
 > > > I'll have to look at that. The observing run was about a year ago, and we did
 > > > not derive the single antenna beam squint, although we have the data and can do
 > > > so. It turned out the BIMA sensitivity was insufficient to do the science
 > > > experiment, so we obtained some test data but did not fully investigate all
 > > > instrumental effects. I will have the grad student who worked with these BIMA
 > > > data look into this and let you know the result.
 > > >
 > > > For spectral line Zeeman work beam squint is the biggest instrumental problem I
 > > > have to deal with, so I always carefully measure the beam squint. An example of
 > > > such a study is described in Crutcher et al., ApJ, 456, 217, 1996; figure 1
 > > > shows a map of the Stokes V beam of the IRAM 30-m telescope. For a synthesis
 > > > observation, where the synthesized beam depends to first order on the UV
 > > > coverage, what we do is make sure that the UV coverage is identical in R and L.
 > > > Primary beam squint can also be important, because of course the synthesized
 > > > beam from the UV coverage is convolved with the primary beam response. But to
 > > > the next level of importance, since the synthesized beam is a small fraction of
 > > > the size of the primary beam, it shows up as a gain difference between the R and
 > > > L synthesized beams that is variable over the primary beam. Such a gain
 > > > difference can be calibrated out straightforwardly. We clearly see the effects
 > > > of primary beam squint when using the VLA for OH and H I Zeeman mapping. I do
 > > > remember that the VLA did obey the standard formula for primary beam squint
 > > > produced by an off-axis feed, but I did this many years ago and don't readily
 > > > have the details. At the next level of importance for Zeeman work, primary beam
 > > > squint will produce a spurious instrumental Stokes V response in synthesis maps
 > > > that can lead to spurious Zeeman detections in the presence of a spatial
 > > > velocity gradient in the spectral line source. This has never been a significant
 > > > problem for my VLA Zeeman work, mainly because for our synthesis mapping we
 > > > track over large ranges in hour angle, and the polarized primary beam pattern
 > > > rotates on the sky and its effects are smeared out. For short observations it
 > > > can be a major problem, however.
 > > >
 > > > Peter Napier wrote:
 > > > >
 > > > > Dick,
 > > > > Dick Plambeck suggested I contact you as one of the few people who have
 > > > > tried to measure Stokes V with BIMA. For ALMA I am looking into the
 > > > > polarization problems caused by laterally displacing the feed from the
 > > > > secondary focus. When you do this you cause beam squint between the RCP
 > > > > and LCP beams which gives rise to an increasing instrumental V term as
 > > > > you move off axis in the primary beam. I would like to check my formula
 > > > > on an existing antenna that has off-axis feeds so I am wondering if you
 > > > > see any evidence of circular polarization beam squint in your BIMA data?
 > > > > Regards,
 > > > > Peter
 > > > > --
 > > > > Peter Napier
 > > > > NRAO, PO Box 0, Socorro NM 87801
 > > > > Ph 505 8357218, Fax 505 8357027
 > > >
 > > > --
 > > > Richard M. Crutcher
 > > > Chair, Department of Astronomy
 > > > University of Illinois
 > > > 1002 W. Green St.
 > > > Urbana, IL  61801
 > > > Voice: 217/333-9581
 > > > Fax: 217/244-7638
 > > 
 > > --
 > > Peter Napier
 > > NRAO, PO Box 0, Socorro NM 87801
 > > Ph 505 8357218, Fax 505 8357027
 > 
 > -- 
 > Richard M. Crutcher
 > Chair, Department of Astronomy
 > University of Illinois
 > 1002 W. Green St.
 > Urbana, IL  61801
 > Voice: 217/333-9581
 > Fax: 217/244-7638
 >