Performance and Reliability of the EVN; September 2008 ====================================== Session 3/2007 -------------- The session consisted of five frequency blocks (L, C, 5cm, 7mm, SX). NME for L, 5cm and C bands were carried out including ftp tests. An ftp experiment (no NME) was carried out at 7mm. The ftp data were processed in a semi-automated way. Two different versions of the schedule were used by EVN and VLBA. This resulted in a mis-overlap of the phase referencing cycle. Some stations reported lots of RFI at L-band. Few stations reported brief data losses due to problems with either Mk5 or FS control computers which required rebooting. This problem seems to occur less frequently than previously reported. Station and correlator feedback for individual stations: Ef - During GB060A, unfortunate choice of LO settings caused problems with phase lock. Problem was solved after a 1.5 hours. The station was correlated as 1-bit station because of low sampler stats. Wb - Gain on BBC01 was reported unstable on a few occasions. Nt - Due to technical problems, the subreflector did not move following the elevation but it was fixed at the 45 degrees position. This degraded the calibration curves of the secondary focus receivers (basically all but the S/X). The 7mm data from Nt are only left pol because of a technical problem discovered just before the session. Furthermore, the receiver was not properly set to the standard frequencies: no fringes were found to Nt during this experiment. Problems during a disk change caused the loss of about 2 hours during EC025B. In EP062, bad sampler stats in upper ~6 subbands (correlated as 1-bit station). On - During GC029, a unknown issue (maybe related to wx) hanged the FS for half hour. Cable wrap (17min lost) during GB062 Tr - The first hour of EH022 was lost in order to restore the correct Mark5a version (this problem was spotted during the ftp test). Phase jumps at 6cm (continued through subsequent sessions; fixed prior April eVLBI.) Hh - During EM067, a thunderstorm leaked into receiver feed and electronics causing loss of the RF signal and high Tsys. Cm - EH022: Receiver in the wrong position for the first 3 hours. Pointing problems during GP044: the telescope control software was sending random offset values. Session 1/2008 -------------- The session consisted of five frequency blocks (L, C, 5cm, K, SX). Two ToO GP046A,B at C-band were observed immediately before and after the session. NME including ftp tests were carried out before starting the observations of each frequency block. The ftp data were processed in a semi-automated way. Some stations reported RFI at L-band. Few stations reported brief data losses due to problems with either Mk5 or FS control computers which required rebooting. The session was heavily affected by the bad weather conditions, 3 experiments abandoned prior to correlation and six other experiments missed either Ef and/or Jb, and perhaps others. Single stations reported several weather related outages (high wind and snow). Station feedback for individual stations (correlator feedback given only for the correlated experiments): Wb - Problems with the power supply of the real time control system before the end of EP061C and the beginning of EP061D. Nt - The station reported "LSB of BBC08 in overflow" for several experiments. In GC030, the cal diode was out of order, no Tsys for the first 3 hours and bbc#8 unlocked for the rest of the observation. Half hour of GB063B lost because the antenna was close to 'prelimit'. Mc - High Tsys due to vacuum receiver problem, during EM059C. On - Problems with the declination motor. The station reported high Tsys in EM058B and EM069C. Tr - Antenna control computer crashed during EL034B. In EA036B, BBC8 and/or BBC7 unlock on and off throughout the experiment. Azimuth drives problem during all the 5cm session. The phase jump problem continued. Cm - Data Throttling during 1Gb/s experiments. Sh - Recording problems (data throttling) during EG037C, most of data lost. Ur - Changes of disk pack in the middle of a scan in most of the experiments. Session 2/2008 -------------- The session consisted of four frequency blocks (L, C, 5cm, K). Two ToO GP046A,B at C-band were observed immediately before and after the session. NME including ftp tests were carried out before starting the observations of each frequency block. The ftp data were processed in a semi-automated way. Some stations reported RFI at L-band. Wb reported particularly bad RFI around 1625MHz. The wind was strong during some experiments at Onsala, which reported severe weather conditions too. Few stations reported brief data losses due to problems with either Mk5 or FS control computers which required rebooting. In EP061E, Nt lost few hours because of a MK5 failure. Some problems also with computers and disk pack: few hours lost at some stations. Station feedback for individual stations (correlator feedback given only for the correlated experiments): EF: Lost some scans for pointings and, in other occasions, wrong focus. In exp EB038B, lost two hours due to computer communication problems. Wb - False storm alarm during exp EO006A: lost 25 minutes. Wrong setup (and few hours lost) during GB064C. TADUmax testing showed reliable data (different apriori clock offsets as a function of SB bandwidth) Nt - The RCP channel of the K-band was broken. Bad sampler stats BBC#1,4,8 (freq. 6030,6035 MHz). Mc - No fringes yet for GS029 (freq. 6030,6035 MHz) On - The station reported high Tsys in the L-band experiments, causes not known. Ro: During the experiment GM064 a series of small problems affect the observations. In particular, pointing problems and low (and unstable) signal is some VC. Cm - Data Throttling during 1Gb/s experiments (cause identified, solution enacted). In exp GS029, wrong LO (no fringes) for part of the experiment. In exp EA038, more Merlin stations were used. Cambridge, Defford, Darnhall and Knockin recorded on the same disk pack. There appears to have been intermittent link failures for Cambridge and Defford, data are ok. Ys - 1st EVN participation in K-band 2/2008 (K, S/X scheduled 3/2008). Giuseppe Cimo' for the Science Operations and Support Group (JIVE)