A NEAR FIELD 3D SCANNER FOR MILLIMETRE WAVELENGTHS P. Schemmel, S. Maccalli, B. Maffei, F. Ozturk, G. Pisano, and M.W. Ng Jodrell Bank Centre for Astrophysics, School of Physics and Astronomy, University of Manchester, Oxford Road, Manchester M13 9PL, UK E-mail: peter.schemmel@postgrad.manchester.ac.uk ABSTRACT Typical millimetre wave astronomical receiver systems make use of quasi-optical (QO) components such as lenses, interference filters, polarisers, or polarisation modulators. Each of these may be located in the near field of additional components. Combinations of these com- ponents lead to standing waves and reflections, which de- grade the overall RF performance. Therefore, it is im- portant to characterise the near field of QO components and QO systems, as well as the evolution of the near field along the propagation axis. Additionally, it is not always feasible to measure the far-field of large systems. This limitation may be overcome by using analytic methods to transform near field measurements into far field radiation patterns. We have developed a 3D near field scanner for millimetre wavelengths. The scanner has a working volume of 50 cm × 50 cm × 50 cm and is coupled to a vector network anal- yser (VNA) operated in the W-band (75-110 GHz). Vol- umetric scanning provides the advantage of being able to directly measure field parameters at all positions in the optical path and to map the evolution of the near field. We present the design and performance of this facility to- gether with the initial near field characterisation of some QO components that are also compared to RF simula- tions. 1. INTRODUCTION Electromagnetic fields generated by radiating antenna are typically divided into three sections, the reactive near field, radiating near field (or Fresnel Region) and the far field (Fraunhofer Region). Although the accepted region definitions have discrete boundaries, the actual fields vary smoothly as they propagate and the distinction between near field and far field regions is not always clear. The re- active near field is defined as the region within a distance of one wavelength (λ) from the antenna. The minimum radial distance from the antenna, at which a wave can be considered in the far field is, r min = 2D 2 λ + λ (1) where D is the largest cross sectional dimension of the antenna [1]. The radiating near field occupies the re- gion between these two boundaries. The composition and form of the radiated field varies significantly between re- gions. The reactive near field is comprised of several plane wave modes in addition to evanescent modes, while the far field is primarily a single plane wave mode [2]. Alternatively, the radiating near field is free from evanes- cent modes, but a superposition of plane wave modes are required to describe the field accurately. If a QO component is introduced into the near field of an- other component, antenna or otherwise, the interaction of the component and radiated field can result in spu- rious reflections and standing waves, which reduce the overall RF performance. Additionally, the several plane wave modes of the near field, interact with the QO com- ponents at various oblique incidence angles [4, 5]. These effects might not be included in some RF models, which can result in significant discrepancies between modelled and measured data. 2. CONCEPT The 3D near field scanner consists of two parallel Z axis and one X axis belt driven rails supported by a 1.0 metre high aluminum frame. A movable carriage is allowed to translate along a Y axis screw driven rail that is mounted to the X axis (Fig. 1). Stepper motors are used to drive the translating rails. The total scanning volume is 50 cm × 50 cm × 50 cm. The scanner has a mechanical posi- tioning accuracy better then 0.1 mm. A Rohde & Schwarz RS ZVA-110 converter, used for the probe, is mounted to the Y axis carriage while a sec- ond converter used for the antenna under test (AUT) is mounted on a separate frame. A RS ZVA-40 VNA is used to measure the S-parameters of the AUT, across the W-Band. Although data is recorded across the entire W-Band, the data in section 4 is comprised of only the