<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
That's pretty weird, Rick. Since it has a behavior that is symmetric
around the meridian, changes in a continuous fashion, and is more or
less common to both polarizations, it could be indicative of
something in the optical path on the antenna that's shifting or
sagging over elevation. A loose subreflector would affect all bands,
and L-band the least. Another possibility is the feed horn itself is
sagging at low elevation. I have no idea how this could happen, but
you did say that crazy ideas were welcome, so that's my
contribution.<br>
<br>
-Wes<br>
<br>
<div class="moz-cite-prefix">On 9/5/2019 2:40 PM, Rick Perley via
evlatests wrote:<br>
</div>
<blockquote type="cite"
cite="mid:e15eb86c-7120-3742-992c-0d36f9b0c74e@nrao.edu"> On
August 13, we observed Cygnus A at L-band and P-band. Two issues
were uncovered. The first (ea28 in RCP) has already been
reported, and a likely cause identified. This note is about the
other -- much more mysterious -- effect.
<br>
<br>
Antenna 13 -- and only this antenna -- demonstrated very
strange changes in both power and correlated visibilities, in all
spectral windows. The behaviors in each SPW are grossly similar,
but quite different in detail. The bad behavior lasted about
three hours, more or less exactly centered on meridian transit.
Similar effects were seen on the calibrator in target source.
<br>
<br>
The overall characateristics are these:
<br>
<br>
1) The total power increased.
<br>
<br>
2) The swtiched power was absolutely stable.
<br>
<br>
3) The visibility amplitude declined.
<br>
<br>
4) The bandpass shapes changed dramatically.
<br>
<br>
5) Each spectral window showed different behavior, but the two
polarizations in a SPW were similar (although not identical).
<br>
<br>
Attached are some plots to illustrate the points:
<br>
<br>
A) EA03-SPW07-SWPower.png
<br>
<br>
Showing the switched power from ea03 -- the antenna adjacent
to ea13. Top to bottom: PDif in RCP, PSum in RCP, PDif in LCP,
PSum in LCP. The color reflects the source: Blue is the nearby
calibrator, green is Cygnus A. Shown in SPW 07 -- 1415 MHz.
<br>
<br>
This is what we expect from a stable system: The PDif panels
(top, and third from the top) are stable. The high noise for
Cygnus A is a result of the increased power from Cygnua A (a
factor of 4 above cold sky). The 2nd and 4th panels show the
total power.
<br>
<br>
All antennas look like this *except* for ea13.
<br>
<br>
B) EA13-SPW07-SWPower.png
<br>
<br>
The same as above, but for ea13. Note the switched power
remains beautifully stable. This shows the receiver itself is not
to blame for what follows. The total power plots (2nd and 4th
panels) show the power smoothly rising, then rapidly dropping to
the level seen at the beginning (and end) of the run. The pattern
is symmetric about transit! Note also that the calibrator powers
(in blue) are also strongly perturbed.
<br>
<br>
C) EA13-SPW08-SwchdPower.png
<br>
<br>
The same as above, but for the next spectral window, SPW08.
Note that the perturbations in power are much less than SPW07.
<br>
<br>
D) EA13-SPW7+8AmpGains.png
<br>
<br>
This shows the calibrator gains for SPWs 7 and 8. A rise
means the visibility flux is too low. Comparing this to the
switched power plots shows that the visibility amplitudes are
dropping more-or-less inversely to the rise in total power. Yet
the gains themselves (as shown by the PDif plots) are unchanged!
This could be a saturation effect, but the rise in the power is no
where near large enough to do this... Note that SPW7 (IF7) and
SPW8 (IF8) have gain variations in opposite directions at Time =
05 hours). SPW07 goes up and down on subsequent scanes, while
SPW08 is reversed.
<br>
<br>
E) EA03-SPW+8AmpGains.png
<br>
<br>
Same as above, but for the adjacent antenna. The gains are
beautifully stable!!! (And the same is seen on the antenna on the
other side of Ea13).
<br>
<br>
F) EA13-BPVar.png
<br>
<br>
Showing the *differential* bandpasses (i.e., changes from the
mean bandpass, averaged over the entire run), for ea13. The
(color-coded) variations are large (+/- 15%),, and very different
between the two IFs. They are similar in shape between
polarization, but different in detail.
<br>
<br>
G) EA03-BPVar.png
<br>
<br>
Same as above, but for the adjacent antenna, EA03. Variations
here are minuscule -- 1%, and entirely due to noise.
<br>
<br>
------------------------------------------------------
<br>
<br>
So what is going on? The only rational explanation I can
think of is a receiver saturation effect. But the saturation must
be out-of-band, since I can't see it in the spectra of any band.
<br>
<br>
But, if a saturation effect, why is it nearly perfectly timed
for meridian transit, and why is *only* ea13 (at W56) affected?
<br>
<br>
All suggestions -- no matter how crazy, will be considered ...
<br>
<br>
Rick
<br>
<br>
<br>
<br>
<fieldset class="mimeAttachmentHeader"></fieldset>
<pre class="moz-quote-pre" wrap="">_______________________________________________
evlatests mailing list
<a class="moz-txt-link-abbreviated" href="mailto:evlatests@listmgr.nrao.edu">evlatests@listmgr.nrao.edu</a>
<a class="moz-txt-link-freetext" href="https://listmgr.nrao.edu/mailman/listinfo/evlatests">https://listmgr.nrao.edu/mailman/listinfo/evlatests</a>
</pre>
</blockquote>
<br>
</body>
</html>