PLANCK is a project of the European Space Agency to be launched in February 2007 by an ArianeV rocket with the Herschel Space Observatory . It is designed for imaging the temperature and polarization anisotropies of the millimetre and submillimetre radiation over the whole sky with unprecedented sensitivity, accuracy and angular resolution using 9 frequency channels ranging between 25 and 1000 GHz. The main source at these frequencies is the Cosmic Microwave Background (CMB), i.e. the radiation emitted by the early universe when, about 300000 years old, ionised hydrogen recombined and became transparent from the visible to radio frequencies of the electromagnetic spectrum. The main goal of the PLANCK mission is to retrieve the main cosmological parameters of the Universe with accuracies of a few percent from the observation and analysis of random small contrast (10–4) features in the CMB. The angular power spectrum of the CMB anisotropies is a function of the fundamental cosmological parameters. A proper measurement of all the angular frequencies of the CMB is essential for an accurate interpretation of the data. In consequence the optical performances of Planck will directly impact the ability of retrieving theses parameters. Recent results of the Willkinson Microwave Anisotropy Probe (WMAP) mission show that polarization information of CMB radiation is very challenging, and that the precise measurement of the CMB could completely change the knowledge we have on our universe (). The focal plane assembly (FPA) of the PLANCK telescope is composed of two instruments. The High Frequency Instrument (HFI) of PLANCK is the most sensitive CMB experiment ever planned (). Together with the Low Frequency Instrument (LFI), this will make a unique tool to measure the full sky and to separate various components of its spectrum. This paper describes the main performances of the HFI beams and compares results obtained with 2 different softwares: GRASP8  and an home-made software developed at the Ireland National University of Maynooth . Specials attention will be paid to polarized beams (100, 143, 217, 353 GHz) and multimoded channels (545 and 857 GHz).
The future ESA space mission Planck Surveyor mission will measure the Cosmic Microwave Background temperature and polarisation anisotropies in a frequency domain comprised between 30GHz and 1THz. On board two instruments, LFI based on HEMT technology and HFI using bolometric detectors. We present the optical solutions adopted for this mission, in particular the focal plane design of HFI, concept which has been applied already to other instruments such as the balloon borne experiment Archeops.
Because of their special properties profiled corrugated waveguide-horn structures are popular as both single-moded and multi-moded bolometer feeds in CMB experiments (e.g. PLANCK, Archeops, QUaD). Although optimised at a spot frequency the horns are usually employed over a relatively wide bandwidth and for single-moded horns the waveguide itself acts as the high pass filter. The horns can be profiled to reduce the horn length and sometimes also a front flared section is added to provide minimum edge taper and spill-over levels (e.g. on PLANCK). In this paper we report on our detailed analysis of the bandwidth properties of such corrugated horns. In the case of multi-moded horns an important issue is how the number of modes varies across the band with the resulting impact on the beam patterns. The so-called "phase centre" location 3 is also an issue. For polarisation selective systems we probe the polarisation purity of the relayed signal across the band and also investigate waveguide details that determine the exact location of band edges. Furthermore any leakage below the expected cut-off will lead to non-idealised cross-polar effects. We have also undertaken a series of laboratory measurements of bandwidth effects in corrugated waveguides to verify the models used.
QUaD is a ground-based high-resolution (up to l ≈ 2500) instrument designed to map the polarisation of the Cosmic Microwave Background and to measure its E-mode and B-mode polarisation power spectra. QUaD comprises a bolometric array receiver (100 and 150 GHz) and re-imaging optics on a 2.6-m Cassegrain telescope 2. It will operate for two years and begin observations in 2005. CMB polarisation measurements will require not only a significant increase in sensitivity over earlier experiments but also a better understanding and control of systematic effects particularly those that contribute to the polarised signal. To this end we have undertaken a comprehensive quasi-optical analysis of the QUaD telescope. In particular we have modelled the effects of diffraction on beam propagation through the system. The corrugated feeds that couple radiation from the telescope to phase-sensitive bolometers need to have good beam symmetry and low sidelobe levels over the required bandwidth. It is especially important that the feed horns preserve the polarisation orientation of the incoming fields. We have used an accurate mode-matching model to design such feed horns. In this paper we present the diffraction analysis of the QUaD front-end optics as well as the electromagnetic design and testing of the QUaD corrugated feeds.
Smooth walled Winston horns have been extensively used as light collectors for bolometric instruments. Used in multimoded operation without a waveguide, the beam shape is top-hat like and a simple equation is sufficient to define its Full Width Half Max. On the other hand, it is well known that corrugated feed horns are more efficient than smooth walled horns and much better well behaved with respect to polarization characteristics and sidelobe rejection. We present in this paper a study of corrugated Winston feed horns which could be used for future Astronomical instruments devoted to Cosmic Microwave Background (CMB) polarization measurements. We show that in this case low cross-polarization can be expected in single moded operation and that they could produce lower sidelobes levels compared to conical or profile shapes.