UNIVERSITY OF CAGLIARI
Physics Department
Complesso universitario di Monserrato
SP Sestu-Monserrato km 0.7
I-09042 Monserrato (CA), Italy

NATIONAL INSTITUTE FOR ASTROPHYSICS
Cagliari Astronomical Observatory
Località Poggio dei Pini
Strada 54
I-09012 Capoterra(CA), Italy

Nichi D'Amico - The first hand-made pulsar backend at Parkes

Modern astronomy, like other branches of science, is continuously developing further complex experiments, so a problem arises when we train students and young scientists: how we can drive them straight to the point of the essential concepts of such sophisticated instrumentation ? In pulsar astronomy, this can be achieved with some simple experiments, in which almost all the technical issues can be very well addressed. Indeed, the construction at the Parkes radiotelescope (see picture on the left) of a simple backend in 1981, represented for me an excellent opportunity to understand the basics of pulsar astronomy. Since then, we developed much more complex experiments, but the basics are always the same.

In our early experiments at Parkers at 1.4 GHz, carried out in 1981 and 1982, Dick Manchester and myself where looking for periodicities of the order of few tens of milliseconds -essentially we where searching new Vela-like pulsars, but at the same time we wanted to get rid of pulse dispersion and scattering which might be rather strong for low Galactic-latitude distant pulsars. The hardware that we developed for such pioneering experiments was hand-made, and was rather simple (click on the picture to enlarge), but it resulted to be successful in the discovery of a first significant sample of short period distant radio pulsars. When a large radio telescope is located close to a University site, as it will be in our case with the Sardinia Radio Telescope, it results that such a simple, easy-to-build, backend can be very useful for training students, up to the point of designing and actually undertake a pulsar observing program.
Simulations showing the effect of Dispersion
Simulation showing the expected response of the adopted antialiasing filter
Schematics of the filterbank connection to the IF section
A 4 x 5 MHz filterbank, and the frequency response of each filter
Electric scheme of the 5 MHz filters