AMTA Paper Archive

The design of hybrid-mode dielectrically-loaded horns [1][2] for antenna test range illumination is described. These horns have a wide operating bandwidth of 5:1 or greater and were designed to replace conventional corrugated- or smooth-walled illumination horns that, typically, have a bandwidth of 2:1or less. Dielectrically-loaded horns have the radiation characteristics desirable for test range illumination: principal plane pattern symmetry, reasonably low cross-polarization and low sidelobes, low reflection coefficient and relatively constant beamwidth. At CSIRO we have developed software and manufacturing techniques to design and make these horns accurately. Measured results, that show close agreement with predicted values, are presented for a horn made for the frequency range, 7 to 40 GHz.

An accurate determination of the radiation pattern coverage is necessary for Flight Termination and Safety (FTS) systems. Vehicles such as the X-34 are physically large and can be difficult to handle and mount for full spherical characterization of the patterns. The question addressed here is "can a partial, full-scale mockup be used for FTS measurement purposes?" Simulations were performed to determine the percent spherical coverage for three candidate X-34 full-scale configurations: 1) Complete mockup, 2) full-length mockup without wings or tail and 3) a partial-length model without wings or tail. The radiation patterns were computed using NEC-BSC and post processing was done to determine the coverage. The FTS UHF percent spherical coverage calculations varied by less than 0.5 dB. For the three configurations, the level at 95% spherical coverage varied from -20.55 to -21.0 dBi for LHCP. Subsequent measurements of case 3) were within 0.5 dB of the values predicted.

For greatest efficiency and accuracy in measuring patterns of a circularly polarized antenna on a planar near field range (NFR), a recommended procedure is to use a fast switched, dual circularly polarized probe. With such equipment one obtains complete pattern and polarization data from a single scan of the antenna aperture. For our task of measuring high gain shaped beam apertures, measurement efficiency is further improved by using a moderately high gain (about 12 dBi) probe that has been accurately calibrated for patterns, polarization, and gain over the test frequency band. Such a probe allows scan data point spacing to be typically at least one wavelength, thus keeping scan time minimized with acceptably small aliasing (data spacing) error. The measured near field amplitude and phase data is transformed via computer to produce the angular spectrum that is further processed to remove the effect of the probe patterns, i.e. probe correction. The final output is a set of (principal and cross) circular­ polarized far field patterns. However on one occasion, due to fast breaking changes in requirements, we were unable to obtain a calibrated circular polarized probe in the available time. For this test we used an available calibrated 12 dBi fast-switched dual linear-polarized probe with software capable of processing principal and cross circular-polarized far field patterns. As anticipated, we found from preliminary tests that the predicted low cross-polarized shaped beam pattern was not achieved when using the calibrated fast Ku band probe switch. Further tests showed the problem to be due to small errors in calibration of the probe switch. This paper will discuss test and analysis details of this problem and methods of solution.

When mounted on spacecraft , pattern of some antennas are perturbed by the presence of satellite body. The prediction of antenna performances including satellite structure effect is generally done at early stage of antenna design but is limited in terms of model complexity. The test on full spacecraft & in far field condition is then necessary. This solution is very expensive as it means for test at satellite level to use Compact antenna Test Range in order to satisfy cleanliness aspects. For the Meteosat Second Generation (MSG) program test on the toroidal antennas need to be performed on different model including a flight model. A good compromise was to use the external omnidirectional antenna range and a part of satellite structure representing the major contributor for the antenna pattern as identified via numerical analysis. The external range offer possibilities that cannot be reached in Compact range, e.g. low cost, full sphere pattern, low frequency range.

Considerable progress has recently been made in the application of phase retrieval methods for phaseless near-field antenna measu rements. These techniques have sufficiently matured so that accurate antenna measurements can be performed when the phase information is either unavailable or inaccurate. A comparison of conventional (amplitude and phase) and phaseless (amplitude only) planar near-field measurements for non-ideal measuring probe locations is examined via simulated array antenna case studies involving both random and systematic errors. It will be demonstrated that the presented phase retrieval algorithm can more accurately reproduce the true pattern of the antenna under test because of the diminished sensitivity of the amplitude of the near­ field, as compared to the phase, with respect to the measuring probe locations. This phase retrieval approach requires no knowledge of the actual measurement locations, other than the nominal location of the two required measurement planes, and is suitable for relatively large probe position errors.

Obtaining far-field radiation patterns of high frequency antennas (>80Ghz) from near-field measurements has been an important issue in the last twenty years. However with frequencies increasing into the millimetre and sub-millimetre bands, questions have been raised about possible limitations on the assessment of such antennas and in particular the measurement of phase. The PTP phase retrieval algorithm addresses the problem by extracting the phase from the knowledge of two amplitude data sets in the near-field. The accuracy of the algorithm is studied by simulation and measurement by means of a numerical/statistical approach. Pseudo-random phase apertures can be generated using Zernike polynomials, which in turn can be used as initial estimates for the algorithm. This paper shows some simulated and measured results for various separations. It can be seen that different pseudo-random phase functions can affect the accuracy of phase retrieved results in particular when the distance between planes is considerably small in relation to the AUT size.

Toshiba Corporation, working with Nearfield Systems Inc., has a fully digital antenna measurement system for digital beam-forming (DBF) antennas. The DBF test facility is integrated with the large 35m x 16m vertical near-field range installed at Toshiba in 1997 [3], and includes the NSI Panther 6500 DBF Receiver as the primary measurement receiver. The DBF system was installed in March 1999 and has been used extensively to test and characterize a number of complex, high performance DBF antennas. A DBF antenna typically incorporates an analog-to­ digital (AID) converter at the IF stage of the transmit/receive (T/R) module. The digital IF signals are transferred to a digital beam-forming computer, which digitally constructs, or forms, the actual antenna pattern, or beams. Since the interfaces to the DBF antenna are all digital, the usual microwave mixers and down-converters are incompatible. The NSI Panther 6500 is designed to interface directly with DBF antennas and allows up to 8 channels of I and Q digital input (16 bits each) with 90 dB dynamic range per channel. The NSI DBF receiver solves the DBF interface problem while providing enhanced performance over conventional microwave instrumentation. [2].

The purpose for this advanced development program was to design, fabricate and test a physically small, broadband, dual-polarized, phased array antenna aperture using Flare Notch elements. The array was designed to operate in the 4 to 18 GHz frequency spectrum, having a VSWR of less than 2:1 and capable of handling 10 watts per element. The array was configured with polarization diversity, essentially, dual cross elements are used which are excited in phase or out of phase depending on the application. One of the significant accomplishments of this research effort was the elimination of grating lobes and the reduction of the size of the elements. Another significant accomplishment is the feeding of dual flare notch elements with a broadband microstrip match network. The antenna elements were implemented using Rogers 4003 materials. Fabrication of the elements and assembly of the elements is being done in a configuration of two rows by twelve elements of which only eight elements are normally excited. The remaining elements are used as parasitics to support the desired radiation pattern. The research work is being done in support of the next generation of solid state broadband radiation systems presently under development for ECM applications.

An important source of error in a compact range antenna pattern measurement is the deviation of the quiet-zone field from the perfectly fiat amplitude and phase of a plane wave field. Although some guidelines and rules of thumb exist that relate the quiet-zone field to the error in the measured antenna patterns, the error or perturbation is dependent on the particular type of antenna that is being measured. For example, the non-ideal quiet­ zone field will produce very different errors for a small horn than for a large phased array. A realistic error budget or uncertainty analysis of the compact-range measurement requires knowledge of the antenna pattern uncertainty as a function of the quiet-zone field and the particular antenna of interest. A simulation method is derived using reciprocity that allows one to quantify the perturbations induced in a given antenna pattern when the quite-zone field distribution is known. This is particularly useful, since one typically has a fair estimate of the antenna pattern and has measured data of the quiet-zone field. The simulation is tested by modelling the antenna as a collection of elemental current sources and simulating the quiet-zone field as generated by elemental current sources. Using this simple simulation model, a closed-form near-field antenna pattern may be calculated for comparison with the more general computer simulation derived from reciprocity.

Many aircraft radome structures are designed to operate simultaneously over multiple RF bands and incident polarizations. Critical parameters must be measured over the electrical apertures of the radome and across each operating band. Automated measurement techniques are required to efficiently collect the large volume of test data required. A modular broadband feed assembly has been developed to allow the simultaneous collection of multi-band, multi-polarization data on a compact range without the need to mechanically change feeds. The feed assembly utilizes a sinuous antenna as the radiating element and is capable of operation from 2-18 GHz with electronically selectable polarization states. Feed design criteria as they relate to compact range antenna and radome measurements are discussed. Of primary importance are reflector illumination pattern, linear polarization cross-polarization level, and circular polarization axial ratio. Polarization switching requirements for a specific test application are defined and the physical implementation of the integrated feed assembly is described. Measured feed and quiet zone performance data is presented for this application. The polarization switching configuration can be readily modified to support other applications.

Nowadays, highly accurate antenna pattern and RCS measurements are performed in compensated compact range test facilities, which fulfil the stringent space requirements for measurements up to 500 GHz and more. As the suppression of diffracted fields from the reflectors mainly determine the quiet zone field performance, the reflector edge treatment is an important design parameter for this type of test facilities. Within the present paper a novel serration design wm be shown. The analyses as well as measurement results exhibit a clear improvement of the quiet zone field performance when compared to previous solutions. The new serration design was implemented and proved with the CCR 20/17 of Astrium GmbH at the Munich University of Applied Sciences.

Compact ranges have found wide use in the pa rametric characterization of high performance radomes such as those found on modern military aircraft. A properly designed compact range facility provides a stable, repeatable test environment suitable for the measurement of small variations in antenna boresight position (beam deflection), antenna pattern distortion, and transmission loss. Radomes have increased in complexity from small structures housing a single antenna to multi-band, multi-system structures large enough to stand inside. Similarly, compact range reflectors have increased in commercial units available today provide quiet zone extents of 12 feet or larger. This paper describes the system design and performance of a compact range test facility designed to test a C-130 Combat Talon II nose radome measuring 7 feet in length and diameter. The facility was constructed at Robins AFB, GA, and is in operation. A description of the facility and its major subsystems is given. Sizing of the chamber and layout of equipment is described. Chamber electromagnetic design considerations are discussed. Electromagnetic design was complicated by the physical size of the structure required to mount the radome, by the fact that multiple antennas and gimbals are present inside the radome during testing, and by the need to use a broad band feed to eliminate mechanical feed changes. Absorber layout and control of spurious reflections is discussed. Electromagnetic performance data is presented.

As the demands on the RF spectrum increase, there is a growing need for antenna test capability at ever higher frequencies. To support our current needs and to accommodate future growth, Ball has outfitted its’ antenna ranges for antenna test from 100MHz through 210GHz. This paper will discuss the considerations and techniques used in extending Ball’s antenna test capability up to 210GHz. The final setup will be discussed and measured pattern data will be presented.

From the earliest days of antenna development, the need for measurement of performance and function has been present. Some characteristics of antennas, such as radiation pattern, are measured by moving one antenna with respect to another. In early antenna testing, outdoor ranges were used to provide a close approximation to the pattern. However, due to the challenges of weather and other environmental effects, antenna testing moved indoors with a number of methods used to compensate for the lack of available space. This paper presents an overview of the history of testing at BATC, from the early days of outdoor testing to the transition to conventional anechoic chambers and nearfield probe facilities. During this time, a variety of techniques have been used to augment standard methods for special requirements, and this paper seeks to communicate some of these methods to the testing community as well as providing a general history of antenna measurement.

A system has been developed for acquiring an antenna’s complete (3D) radiation pattern and radar cross-section (RCS) measurements. The system consists of a motion controller, a network analyser and tower assembly. The tower assembly is in an anechoic chamber. The tower has a novel design. It uses three motors in a special configuration, thereby allowing 2 ½ degrees of freedom. This freedom gives the ability to run complete antenna or RCS measurements automatically. Another advantage stemming from the degrees of freedom is expansion of the range of measurements. This is enabled by a variety of possible positions inside the chamber. Tests have also been carried out on system performance. The data acquisition rate becomes crucial when dealing with 3D pattern measurements. The performance of an HP 8720 or 8753 network analyser series can be dramatically increased by using the power sweep mode for data acquisition. Together with the “external trigger-on-point” mode, this gives the best positioning accuracy. The six-month experience has demonstrated the flexibility and reliability of the set up and ideas.

It is a very common practice to over specify the Quiet Zone performance requirements for an anechoic chamber. Very often what is done is a person who is in need of a chamber contacts someone with a similar facility, often a supplier or a customer, and simply patterns their performance requirement after what the other guy has done. This often results in a chamber, which is specified to a tighter performance requirement than is actually needed to perform the particular measurements required and can cost thousands of dollars more than is necessary. Qualcomm had a requirement to build a chamber for the evaluation of various antenna designs for mobile communication equipment. Due to building and space limitations the “ideal” size for a chamber operating in the 800 Mhz to 6.0 Ghz was not available. Qualcomm worked with AEMI to define the performance parameters to provide them with the best performing chamber that could be built within the restricted space available. Once the design parameters were defined adequately the chamber deign was developed and the chamber was built. Once the chamber was built Qualcomm went about defining the best test methods and parameters that could be achieved given the performance limitations that were evident in the design due to the compromises that had to be made in the limited space available to accommodate the chamber. This paper will discuss the design process, the design limitations and the methods used to overcome the performance compromises made in the development of the chamber and its intended purpose.

While there are standard test methods to characterize the performance of passive antennas, active antennas (with integrated amplifiers) and more complex systems with adaptive functionality create new testing challenges, both in definition and approach. Active antenna gain is a combination of the antenna gain and the embedded amplifier gain. Since these amplifiers may be distributed throughout the array with gain variations between amplifiers, there is a challenge in performing measurements that separate the two gain components. For adaptive antennas, the pattern changes with the incident angle of the test signal, so the adaptive function is often disabled to provide a snapshot of the system, like antenna patterns, for a particular set of conditions. In other cases of adaptive antennas, the composite system performance is measured for angular changes while the system adapts. This paper presents an overview of the testing of both active antennas and adaptive antenna combining systems. Examples of the types of test metrics and errors will be given.

A usual way of performing pattern-measurements on circularly polarized antennas is by measuring the linear components of the field and mathematically converting those to the left-hand and right-hand circular components. These synthesized circular components are sensitive for a number of factors: The exact orthogonality of the measured linear components, the measurement-accuracy of both phase and amplitude of the measured linear components, the polarization-pureness (or the accuracy of the description of the polarization-characteristics) of the probe, etc. This paper analyzes these factors, using a computer-model. An indication on the requirements to be imposed on the measurement-equipment is provided.

Holographic back-projections of planar near-field measurements to a plane have been available for some time. It is also straightforward to produce a hologram from cylindrical measurements to another cylindrical surface and from spherical measurements to another spherical surface1-7. In many cases the AUT is approximately a planar structure and it is desirable to calculate the hologram on a planar surface from cylindrical or spherical near-field or far-field measurements. This paper will describe a recently developed spherical hologram calculation where the farfield pattern can be projected on any plane by specifying the normal to the plane. The resulting hologram shows details of the radiating antenna as well as the energy scattered from the supporting structure. Since the hologram is derived from pattern data over a complete hemisphere, it generally shows more detail than holograms from planar measurements made at the same separation distance.

Future scientific and earth observation instruments as MASTER, PLANCK and HERSCHEL of ESA/ESTEC are working in the sub-millimeter wave range. For measurement of the instruments, a study named ADMIRALS was performed, mainly to identify the most suitable test facility, procure transmit and receive modules and perform measurements up to 500 GHz. The CCR 75/60 of Astrium GmbH, Ottobrunn, was selected for the facility calibration and the pattern verification with an Representative Test Object (RTO). The measurements were performed in three different frequency bands between 200 and 500 GHz. The mmwave transmit and receive modules were designed, manufactured and tested by Radiometer Physics GmbH (RPG). A cost efficient design was achieved by a modular concept. Within this paper, the design and realization of the modules as well as most characteristic performance parameter will be presented.