[evlatests] VLA image -- and an interesting clue ...

[George Moellenbrock]

Michael, et al.,

To answer a few of your (still-relevant) questions from this morning:

>> smaller.  Closely spaced antennas show very similar phase fluctuations, as
>> expected.
>
> I assume these fast variations are seen on the simultaneous VLA data as
> well? Maybe Rick can comment.
>

There is nothing unusual about this, I think, but it does call into question the
veracity of the calibration of the blank field.  Interestingly (or not, in light
of new info re antenna 28!), I was briefly startled by the fact that
ants 2 and 28
show virtually the same post-bandpass phase(t) (offsets and fluctuations)
relative to any other antenna---equivalently, the 2-28 baseline has ~zero
phase throughout, with no fluctuations (except noise).  I briefly pondered
the possibility that the correlator was fed the ~same bits for these two
antennas.  I suppose this is pretty unlikely, and there would have been
other more obvious clues, of course.  The offset phase similarity is
more likely just a coincidence (and the same fluctuations are seen at
other nearby antennas, at different offsets).   If antenna 28
has its phase somehow garbled, is it interesting that it comes out the ~same
as antenna 2?



> This is related to a point Ken has often made: the residual delay clunking
> we see with WIDAR is much less than the residual delay clunking with the
> VLA correlator, where we don't notice them because the dump times are long
> and the frequency resolution is poor.  If delay clunking were a killer we
> would have much worse problems with the VLA correlator.  Ken, could you
> send out a brief summary of this for the VLA compared to the WIDAR
> correlator?
> I think at this point that would be useful to have written down for all of
> us
> to ponder.
>

Right, and the bandwidth is always less than 50 MHz, especially when
one has the resolution (but less per-channel sensitivity) to discern the
clunking.  Presumably, the notorious BLCAL is routinely compensating
for this effect, which kicks in by ~10000:1 for the widest VLA bandwidths.

> - Is the change in dynamic range due to increasing noise or decreasing
>  signal?

Hard to say.  0217 is nominally pegged to 1.0 across the band by the
calibration, so this would tend to hide changes in signal.  With ants 25, 28
removed, I've reimaged my 0217 on-source cube, and I still see an
interesting decline in dynamic range away from the central channel.
See attached plot which shows the (normalized) peak, the rms and
the ratio.  The shape of the ratio is driven mainly by the rms here.
(I show 32 2MHz channels in the central 50% of the 1436 MHz subband.)

At some point, I'll see if I can recover the highest dynamic range in
a different
channel by registering the calibration somewhere other than the band
center.  If this works, then it exonerates pure noise variations with channel,
and points to something systematic with the signal's survival through
the system, e.g., delays, etc.

-George
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SnrVsChan.png
Type: image/png
Size: 65621 bytes
Desc: not available
URL: <http://listmgr.nrao.edu/pipermail/evlatests/attachments/20090714/e4f00a9e/attachment.png>