Astron. Astrophys. 322, 846-856 (1997)

3. Observations

For reasons detailed later, most of the analyzed profiles were obtained within a short period of about three weeks in Summer 1994, using the 100m-radiotelescope of the MPIfR. In general, we observed at five different frequencies between 1.4 GHz and 32 GHz presented in Table 1, whereas all but the 1.4 GHz observations were made with receivers installed in the secondary focus of the telescope. The signals were detected across the full bandwidth and digitized by 10-MHz V/f-converters. The data were sampled and folded synchronously with the topocentric pulse period before a sub-integration of about 15s was transferred to disk. Pulse profiles at 2.25 GHz were measured simultaneously with data at 8.5 GHz using a dual frequency system. For this frequency only signals of one circular polarisation were available, while all other profiles represent total power data. A calibration signal of an external noise diode was switchable synchronously to the pulse period and ensured that possible gain differences were removed while creating total power profiles. Further details about the observing system are given by Seiradakis et al. (1995) or Kramer (1995).

Table 1. Summary of observations. Besides the observing frequency and the epoch of observation, we quote the used bandwidth, the typical system temperature, , the gain of the telescope, G, and the type of the analyzed signals. Total power signals were obtained by adding gain-corrected LHC and RHC signals.

For some pulsars, we include profiles obtained at 1.41 GHz, although these profiles were generally observed at a later epoch. Observations at this low frequency also involved an incoherent pulse de-disperser, consisting of four 60 0.667 kHz filterbanks and a digital delay line. Its construction principle causes a small additional time delay of the signals due to the rise time of the built-in narrow filters and the delay line used for the on-line de-dispersing. Although these offsets in the times-of-arrival (TOAs) can be corrected for, we use such filterbank data only in those cases where the precision of the determined TOA is not affected by these offsets anyway, or where additional physical insight is gained. The profiles at all other frequencies were measured by detecting narrow or broad band signals directly (see Table 1), so that significant instrumental delays can be excluded. If we include profiles obtained at epochs different from those given in the Table 1, it is explicitly stated during the discussion of the results.

© European Southern Observatory (ESO) 1997

Online publication: June 5, 1998

helpdesk.link@springer.de