[evlatests] Some off-center imaging tests/problems

[George Moellenbrock]

Hi Ed-

Re calibrating the delay in CASA...

How are you trying to do this?  BPOLY?   More information please.

I have found that an ordinary B solution does fine if you have sufficient
SNR/channel.

BPOLY cannot (yet) cope with cycle slips within the band.
However, I've found BPOLY's 1st order term detects the correct delay
(in some peculiar units) if the delay is small enough and has
no slips.

CASA does not care what the frequency order of the channels
is.  (Definitely true for sampled B; I am mostly sure for BPOLY.)

By early August, I should have a CLCOR-like task that will
permit manually correcting big delays.  I am also tinkering
with a delay detector within plotms (transform the padded
visibility spectrum and find the peak) that one can use
to plot baseline delays, and from which values for
feeding the CLCOR-like task can be determined.

This delay detector is, of course, the sort of front end
needed to make BPOLY handle arbitrarily large (within
reason) delays, too.....

-George


On Mon, Jul 13, 2009 at 5:49 PM, Ed Fomalont<efomalon at nrao.edu> wrote:
> Hello all,
>
> DATA BASE:
>
> ---I've been looking at the data base:  X0217O1a.55024.720578472225.ms
>
> ---2 hours at x-band:  256 channels in 4 IF's, at frequencies
>
>      Frequencies:   8.4798,  8.3518, 7.9678, 7.8398
>      Channel separation = -0.5 MHz
>
>      THE CONSECUTIVE FREQUENCY CHANNELS GO DOWN in FREQUENCY!
>
>   egs.  IF1, channel 1 freq = 8.54375 GHz, channel 127.5 = 8.47975
>
>      WAS THIS THE INTENDED SETUP?
>
> --- 1/3 of the time 0217+738 at phase center
> --- 2/3 of the time 0217+738 offset 60" to the east,
>    interleved.
>
> REDUCTIONS:
>
> --- Moved to AIPS via casa exportuvfits (also tried CASA)
>
> --- Standard editing, calibration, in AIPS.
>
>    Most antennas had a large delay giving 1 to 2 turns over each IF.
>    Very little flagging needed, good amplitude, phase stability
>    based on on-axis data.  I did the same reduction for CASA.  But,
>    somehow the large delay in each IF confused CASA so that the
>    calibrated data (applycal) was not usable.  I'll follow this up
>    later.
>
> --- Image of 0217+738 looks good.  1000:1 dynamic range.  Should be
>    much better, but this is a detail for now.  Normalized to 1 Jy.
>    Since this source was self-caled, this is not surprising.
>
>    ATTACHED source.pdf
>
> --- After applying 0217+738 calibration, image of 0217/offset very
>    poor.  After changing the sign of the phases, at least this
>    image is centered where it should be, but ...
>
>    ATTACHED offset.pdf
>
> --- In order to compare 0217/offset data with what is expected, I
>    made a model data set of 1 Jy, 60" offset from the phase center
>    using the same data coverage.
>    This image is not as good as that of the real source at the
>    phase center.  It should be better since it has no NOISE.  Hence,
>    there may be slight errors in the data base effecting the phase
>    that task UVMOD has calculated.
>
>    ATTACHED model.pdf
>
> --- Compared phases of 0217/offset with phases associated with the
>    model, a source 60" from the phase center.
>
>    ATTACHED PHASE.pdf
>    yellow = model phases
>    blue = observed phases
>
> --- Used EA02 as antenna reference.  There is good agreement between
>    model phases and measured phases for all antennas
>    EXCEPT FOR EA24 and EA28???
>
> --- Application of the blue phase to the offset data base produced a
>    good quality image at the phase center.  Hence, there is no
>    indication of closure errors.
>
>
> BOTTOM LINE:
>
> --Main point:
>
>    Calibrated phases for the offset source for eight out of
>    ten antennas source look good!   The general amplitude and
>    phase stability over the two hours was excellent with very
>    little editing needed (quack and a bad record near the beginning)
>
>    BUT, EA24 and EA28 have slowly changing large phases offsets for
>    the offset position, but were calibrated to zero for the onaxis
>    source.  So, this is a differential problem.
>    Any technical reasons for this, for example a large instrumental
>    phase change versus antenna pointing?  This could be caused by a
>    large antenna position error for these two telescopes of ~60 meters
>    (see below) These offsets are sufficient to produce the crummy
>    image.
>
> --Secondary points that need follow-up
>
>    a. The model source of perfect data is not as good as the real
>       source model.  Somewhere, the aips task UVMOD is getting
>       somewhat incorrect information on how to move a source 60".
>       Will check antenna file and other methods like uvfix.  However,
>       this might be consistent with a large antenna-offset error,
>       causing the model phases to be in error.  To produce an offset
>       of 180 deg phase for a position change of 60" requires an
>       antenna error of ~1700 wavelengths, or only 60 meters.  Check
>       that the correlator has not changed these two antenna positions.
>
>    b. CASA had problems calibrating the large delay error.  Both the
>       bandpass and gain calibrations were in good agreement with the
>       AIPS cals, but somehow applycal couldn't calibrate the data
>       properly.  Needs more investigation.
>       Perhaps, the negative increment of frequency/channel is a
>       problem.
>
>
> Cheers,  Ed
>
>
> _______________________________________________
> evlatests mailing list
> evlatests at listmgr.cv.nrao.edu
> http://listmgr.cv.nrao.edu/mailman/listinfo/evlatests
>
>