Dear PI, We have examined the data for project RDV53 on 28 Sep. 2005. The contact person for this project was Craig Walker. Here's a summary: 1. Tape weights vs. time plots have been generated for the entire time range of your experiment. These are a measure of tape record/playback quality, representing the fraction of valid data samples. Data with weights below 70-75% should be flagged. However, you may want to be more cautious when dealing with non-VLBA stations. The easiest way to estimate the best weight threshold is by looking at the tape weights vs. time plots generated here. You can find the weights plots at /home/vlbiobs/astronomy/sep05/rdv53/sniffer/final/wtsfile.ps. If your experiment involves more than one distribution tape then there will be a tape# subdirectory between /final and wtsfile.ps. See /home/vlbiobs/README.sniffer for instructions on how to interpret this plot. 2. Delay, rate, phase, and amplitude plots were made for the observation. 3. Autocorrelation bandpass plots were generated for all antennas for scans on all sources. 4. Cross-correlation bandpass plots were generated for all baselines to FD, KK, ON for scans on all sources. 5. Gzipped PostScript plots of Tsys and other monitor data for each available VLBA antenna can be found at /home/vlbiobs/astronomy/sep05/rdv53. 6. The jobs associated with the correlation of RDV53 can be found at: /home/vlbiobs/astronomy/sep05/rdv53/jobs. These files provided the correlator with all ancillary data needed for VLBI, including: correlation parameters and telescopes correlated in the final production. The job numbers are: 1520-1549 7. Operations staff have devised a web browser to navigate the file server vlbiobs, as well as view and retrieve its text and PostScript files, gzipped or not. This browser can be reached through the VLBA homepage http://www.aoc.nrao.edu/vlba/html/VLBA.html, initially under Data Analyst but then under Aspen. All of the files mentioned above can be accessed either with this browser or by FTPing to the vlbiobs account on vlbiobs. Automated Calibration Transfer for VLBA Correlator Output --------------------------------------------------------- The first phase of automated calibration transfer for data from the VLBA correlator has been completed, and was used for your observation. This transfer of calibration information includes data from the 10 VLBA antennas, as well as selected information from the VLA and Effelsberg, which currently provide VLBA-style monitor data. Significant changes to AIPS have been required to introduce calibration transfer, so users must have the patched version of 15OCT98 AIPS, or any later version, beginning with 15APR99. Help files for a number of AIPS tasks have been updated to reflect the new calibration procedures. There also is a new version of the VLBI chapter of the AIPS cookbook, available from http://www.cv.nrao.edu/aips/aipsdoc.html, that includes more details on how to cope with the calibration transfer process. The calibration-transfer process relieves observers of the burden of creating and inputting calibration files for VLBA antennas. Instead, this information is now provided as tables attached to the FITS data sets output by the VLBA correlator. The ancillary data include antenna gain (GC table), system temperature (TY table), pulse calibration (PC table), flags (FG table), and weather (WX table). The wise observer will not modify these original tables; processing errors might then force the data to be reloaded using FITLD. See the description of MERGECAL in Section 9.2.1.7 of the new cookbook chapter for more detail. Of course, skeptical users can simply delete the appropriate tables created by FITLD and generate their tables in the old manner. Phase 2 of calibration transfer will include supply of data from more external telescopes, and probably will proceed incrementally, depending on both the availability of the external information and the implementation of new software in Socorro. At present, ancillary data from most external telescopes must still be loaded in the old manner, and observations of strong sources may be needed for manual pulse calibration at those telescopes. Up-to-date instructions on coping with observations including external telescopes can be found at http://www.nrao.edu/vlba/html/OBSERVING/cal-transfer/cal-transfer.html. Please send comments on calibration transfer to julvesta@nrao.edu, and send bug reports to daip@nrao.edu, with a copy to julvesta@nrao.edu. NOTES: The playback quality was generally good at all stations. Autocorrelation bandpass plots demonstrated siginificant RFI in S band prominent at some stations. AC plots are included in the package. The new regimen of "blind" correlation using calculated values for clocks with an increased number of spectral channels is working well. Any delays in the processing were unrelated to clock searches. ON: Disk pack internal ID lacked a "-" character so Mk5 system did not recognize it as a disk. Walter Brisken logged into Mk5 and added "-" to VSN. We correlated ON with a residual delay near -1 micro second since we now correlate "blind" using clocks calculated from gps data and historical offsets. The reason for this unaccounted delay is currently not known, but Craig Walker is looking into it so we can correct for it prior to correlating ON again. With the new parameter of 64 spectral channels used for the correlation, fringing is well within the window. GC: We were advised of an offset of approximately 300 nano seconds additional to the gps data, but decided not to apply it so as not to compound the error if we misapplied the sign. This yielded a delay average offset that was 124 nano seconds with respect to KK. Stats: Under the new regimen, no clock searches are perform, so no time is used for that purpose during processing. The total correlation hours used was 22h58m including jobs re-run for whatever reason. The release date for RDV53 was 27Oct2005.