NGC3079: Exgal Formaldehyde VLA Observations

TIP Last Modified: JeffMangum - 15 February 2012


  • Observations made on 2011-09-30 were setup to use a correlator integration time of 5 seconds. Due to an error in the WIDAR correlator, though, these measurements were actually acquired with an integration time of 1 second. Here is the message sent to affected observers:

Between the dates of September 20 23:00 UT and December 3 00:00 UT a bug in the EVLA's Correlator Back End (CBE) computer resulted in only 1 second's worth of data being saved per integration, no matter what integration time was selected in the observing script.  This could obviously have a serious impact on the sensitivity you will achieve for your project 11B-026, and we are making every attempt to repeat the most severely affected priority A scheduling blocks before the end of the D configuration.  We will be extending the D configuration by 1 week to accommodate this make-up time.

While we are undergoing commissioning of the EVLA we would appreciate it if all users could inspect their data as soon as possible and report any problems through the NRAO Helpdesk, at

Claire Chandler, Head, Array Science Center

Dale Frail, Assistant Director for NM Operations


  • Observations were re-done on 2011-12-12.
  • Two minor errors were made in setting-up these observations:
    • I neglected to note that there are no overlap channels between correlator subbands. This means that there are "gaps" between subbands where the noise will be larger than that away from these boundaries. One of these subband boundaries is exactly in the middle of the total band observed. All observations of NGC3079 were made with the line placed exactly in the center of the band, placing it on a subband boundary. Analysis of the data acquired 2011-09-30 suggests that the excess noise at these boundaries is not too great (about 20%).
    • I used a velocity resolution of about 10 km/s, which is a bit too course. Should have used 5 km/s. 2011-09-30 measurements suggest that the line is still spectrally resolved, but just barely.


See my listing of the CASA analysis commands used and useful output from my analysis of the 2011-09-30 measurements. Should be a useful guide to the analysis of other Ku- and C-band measurements.

I have posted a cleaned image cube which resulted from this analysis. Note that these data need to be combined with the subsequent re-observation of NGC3079 as this image cube contains only 1/5 of the data that we were supposed to get (correlator integration time error). At least this image cube shows that there is an absorption line detected (pretty much a point source) at the same intensity as we measured with the GBT.

-- JeffMangum - 2012-02-10
Topic revision: r2 - 2012-02-15, JeffMangum
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding NRAO Public Wiki? Send feedback