Carefully conducted polarimetric observations still suffer from polarization state changes due to gain and phase delays, cross-coupling, and depolarization in the observing setup. A detailed analysis is presented here for an adding and a multiplying polarimeter used with three different receivers with the Effelsberg radio telescope to perform high quality polarimetry. The analysis consists of a method to dynamically detect and account for the differential gains in the observing system. The gain monitoring is based on the injection of a calibration signal prior to the polarizer. This signal is synchronous to the pulsar period and is dynamically adjusted for imbalances between the polarization channels if needed. Once gain corrected data are available, a detailed procedure is followed to account for cross-coupling effects. These effects are then removed from the measurements. Using the method we have described, the performance of the observing system can be routinely evaluated. This reduction procedure provided evidence for the stability of the polarization characteristics of the instrumental set-up used, and ensured the high quality of the polarization data. The sample of 28 pulsars at many frequencies presented here was also used to derive emission altitudes based on a method developed by Blaskiewicz et al. (1991) and the results and interpretation are presented in von Hoensbroech & Xilouris (1997). The polarization data presented here comprise a reliable database at cm-wavelengths which can facilitate further studies of polarization effects of pulsars at high frequencies.
Acknowledgements
We want to express our gratitude to Drs. M. Kramer, A. Jessner and R. Wielebinski for their constant help, support and valuable advice. We are in particular grateful to C. Salter for extremely helpful suggestions. Arecibo Observatory is operated by Cornell University under cooperative agreement with NSF.
Figure 5: PSR B0301+57 and PSR B0301+19.
For each profile two panels are plotted. The upper one shows total power (solid),
linearly- (dashed) and circularly [] (dotted) polarized intensity.
The lower plot shows the derived PPA-points with their
error bars. In all cases only such PPA-points are shown which have
a signal-to-noise ratio above a certain level, depending on the
polarisation of the pulsar.
In those cases when a fit of the RVM to the data was applied, the
fitted function is plotted. The points which were considered for
the fits are marked with circles
Figure 18: PSR B1237+25 and PSR B1642-03
Figure 19: PSR B1822-09 and PSR B1915+13
Figure 21: PSR B1946+35, PSR B2016+28 and PSR B2020+28
Figure 23: PSR B2054-16, PSR B2154+40 and PSR B2310+42
Figure 24: PSR B2319+60 and PSR B2351+61