[evlatests] VLA/EVLA L-band noise, aliasing

[Gustaaf van Moorsel]

The narrow band (0.78MHz) L-band results I sent out on Monday were
affected by the timing problem which hit all observations over a four
day period including the past weekend.  With the timing problem gone,
I repeated the tests.  The results are much more encouraging.

The attached plot shows the rms across the band in images of an empty
field.  This is done in four cases: 1 - all antennas, 2 - VLA - VLA
baselines, 3 - EVLA - EVLA baselines, and 4 - VLA - EVLA baselines.
Calibration was done with the usual precautions against aliasing:
flag all EVLA - EVLA baselines prior to calibration, and put them
back in before imaging.  All rms values were normalized to account
for the different number of visibilities in the four cases.

Now that the EVLA - EVLA baselines no longer suffer from the enormous
closure errors introduced by the timing error, the rms on EVLA - EVLA
baselines is only 11% higher than on VLA - VLA baselines (for the upper
end of the band where the effects of aliasing are small).  The observ-
ation took place at elevations between 40 and 50 degrees to minimize
the influence of elevation on VLA and EVLA sensitivities.

Gustaaf
-------------- next part --------------
A non-text attachment was scrubbed...
Name: L0320.pdf
Type: application/pdf
Size: 22042 bytes
Desc: not available
URL: <http://listmgr.nrao.edu/pipermail/evlatests/attachments/20080110/48394c53/attachment.pdf>