An e-VLBI science run took place on the 28th of March 2007, involving Jb (MarkII), Cm, On, Wb, Tr and Mc, at a data rate of 256Mbps. This was the first science run in the L-band, and the first to be completely conducted by an operator and a support scientist, without the direct supervision of the technical staff. This illustrates the increase in robustness and user-friendliness of the system over the past few months.
Because the science observations started very early on the 28th, it was necessary to do the clock searching during the technical test run on the 27th. In addition, a number of problems were fixed. A quick analysis of the data showed that half of the Cm signal was missing. The explanation: due to the microwave link limitations Cm can do two 8 MHz subbands in both polarizations. These were placed in the central subbands. The e-VLBI schedule is always 512 Mbps, and we adjust to the highest possible bitrate during the run - which is currently 256 Mbps, thus leaving out half of the signal. The Cm signal was moved to BBcs 3-4 the next day. Some very strong phasecals were also seen in the Westerbork data, in a few of the subbands; for this reason Wb phasecals were switched off for the science run.
The next day the experiment started smoothly at 4:00 UT. There was one hour to start up things before moving to the science targets at 5:00 UT. There were no problems till 6:30 UT, when the correlator had to be restarted several times because of a crash of the TSPU unit (responsible for the timing of the correlator). Till 10:30 UT the experiment went fine, with occasional job restarts due to either Mark5 or Station Unit problems. The Mark5 at JIVE connected to Wb Westerbork experienced difficulties with time synchronization a couple of times, mostly after gaps in the schedule. This issue will have to be investigated further. The remaining two and half hours did not go very well, unfortunately. Many correlation job restarts were needed due to various problems with Station Units and correlator, which were not directly e-VLBI related.
The co-I of the science proposal, Javier Moldon, was at JIVE to follow the observation, and to reduce the data. The amplitude calibration files from the telescopes were promptly received, which was very much appreciated. The amplitude corrections during the data analysis were less than 10% (typically 4-7%) in all case, except for Cambridge - because Tsys measurements cannot be done over the microwave link. The Cm amplitudes were scaled in the imaging/self-calibration phase without problems.